CVPR-2024-Papers

官网链接：https://cvpr.thecvf.com/

研讨会 :bell:：6 月 17-18 日

主会 :bell:：6 月 19-21 日

历年综述论文分类汇总戳这里↘️CV-Surveys施工中~~~~~~~~~~

2024 年论文分类汇总戳这里

↘️WACV-2024-Papers ↘️CVPR-2024-Papers ↘️ECCV-2024-Papers

2023 年论文分类汇总戳这里

↘️CVPR-2023-Papers ↘️WACV-2023-Papers ↘️ICCV-2023-Papers

2022 年论文分类汇总戳这里

2021 年论文分类汇总戳这里

2020 年论文分类汇总戳这里

💥💥💥收录论文已全部更新，并全部分类完成！！！

🏆Best Papers

• Generative Image Dynamics
  :house:project
• Rich Human Feedback for Text-to-Image Generation

🏅Best Paper Runners-Up

• EventPS: Real-Time Photometric Stereo Using an Event Camera
• pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction

🥇Best Student Papers

• Mip-Splatting: Alias-free 3D Gaussian Splatting
  :star:code
  :house:project
• BioCLIP: A Vision Foundation Model for the Tree of Life
  :star:code

🥈Best Student Paper Runner-Ups

• SpiderMatch: 3D Shape Matching with Global Optimality and Geometric Consistency
• Image Processing GNN: Breaking Rigidity in Super-Resolution
• Objects as Volumes: A Stochastic Geometry View of Opaque Solids
• Comparing the Decision-Making Mechanisms by Transformers and CNNs via Explanation Methods

目录

:cat:	:dog:	:tiger:	:wolf:
1.其它(other)	2.Image Segmentation(图像分割)	3.Image Classification(图像分类)	4.Image/Video Super-Resolution(图像超分辨率)
5.Image/Video Compression(图像/视频压缩)	6.Image/Video Captioning(图像/视频字幕)	7.Image Progress(图像处理)	8.Image Synthesis(图像生成)
9.Face(人脸)	10.Medical Image Progress(医学影响处理)	11.3D	12.Video
13.HPE(人体姿态估计)	14.HAR(人体动作识别检测)	15.Object Detection(目标检测)	16.Point Cloud(点云)
17.Automated Driving(自动驾驶)	18.SLAM/AR/VR/Robotics(增强/虚拟现实/机器人)(机器人)	19.Object Pose Estimation(物体姿态估计)	20.Optical Flow Estimation(光流估计)
21.Few/Zero-Shot Learning/DG/A(小/零样本/域泛化/域适应)	22.Deepfake Detection	23.Sound(语音处理)	24.ML(机器学习)
25.Object Tracking(目标跟踪)	26.Information Security(信息安全)	27.Vision-Language(视觉语言)	28.UAV/Remote Sensing/Satellite Image(无人机/遥感/卫星图像)
29.MC/KD/Pruning(模型压缩/知识蒸馏/剪枝)	30.Person Re-Id(人员重识别)	31.Edge Detection(边缘检测)	32.NLP(自然语言处理)
33.NeRF	34.Human–Computer Interaction(人机交互)	35.Scene Understanding(场景理解)	36.4D Reconstruction(4D 重建)
37.OCR	38.VQA(视觉问答)	39.Motion Generation(动作生成)	40.Scene Graph Generation(场景图生成)
41.Graph Generative Network(GNN/GCN)	42.Image Retrieval(图像检索)	43.Image Matching(图像匹配)	44.Image Fusion(图像融合)
45.NAS(神经架构搜索)	46.Industrial Anomaly Detection(工业缺陷检测)	47.Dense Predictions(密集预测)	48.Semi/self-supervised learning(半/自监督)
49.Dataset(数据集)	50.OOD Detection	51.Style Transfer(风格迁移)	52.Biomedical
53.Light-Field(光场)	54.ViT	55.REC(指代表达理解)	56.Visual emotion recognition(视觉情绪识别)
57.Visual Relationship Detection(视觉关系检测)	58.Fisheye Images(鱼眼图像)	59.Clustering(聚类)	60.Sketch(草图)
61.Gaze	62.全家桶

62.全家桶

• UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio Video Point Cloud Time-Series and Image Recognition
  :star:code用于音频、视频、点云、时间序列和图像识别的通用感知大内核卷积网络
• GPT4Point: A Unified Framework for Point-Language Understanding and Generation点语言理解和生成的统一框架
• AvatarGPT: All-in-One Framework for Motion Understanding Planning Generation and Beyond用于运动理解、规划、生成等的一体化框架

61.Gaze

• Sharingan: A Transformer Architecture for Multi-Person Gaze Following目光跟随
• From Feature to Gaze: A Generalizable Replacement of Linear Layer for Gaze Estimation

60.Sketch(草图)

• What Sketch Explainability Really Means for Downstream Tasks
• SketchINR: A First Look into Sketches as Implicit Neural Representations
• Open Vocabulary Semantic Scene Sketch Understanding草图理解
• CAD-SIGNet: CAD Language Inference from Point Clouds using Layer-wise Sketch Instance Guided Attention

59.Clustering(聚类)

• MoDE: CLIP Data Experts via Clustering聚类
• Fine-Grained Bipartite Concept Factorization for Clustering
• 多视图聚类
  • Investigating and Mitigating the Side Effects of Noisy Views for Self-Supervised Clustering Algorithms in Practical Multi-View Scenarios
  • Learn from View Correlation: An Anchor Enhancement Strategy for Multi-view Clustering
  • Differentiable Information Bottleneck for Deterministic Multi-view Clustering

58.Fisheye Images(鱼眼图像)

• Deep Single Image Camera Calibration by Heatmap Regression to Recover Fisheye Images Under Manhattan World Assumption鱼眼图像

57.Visual Relationship Detection(视觉关系检测)

• Groupwise Query Specialization and Quality-Aware Multi-Assignment for Transformer-based Visual Relationship Detection
  :star:code

56.Visual emotion recognition(视觉情绪识别)

• EmoVIT: Revolutionizing Emotion Insights with Visual Instruction Tuning
  :star:code视觉情感理解
• 多模态意图识别
  • Contextual Augmented Global Contrast for Multimodal Intent Recognition

55.Referring Expression Comprehension(指代表达理解)

• ScanFormer: Referring Expression Comprehension by Iteratively Scanning
• Zero-shot Referring Expression Comprehension via Structural Similarity Between Images and Captions
  :star:code零样本指代表达理解
• Revisiting Counterfactual Problems in Referring Expression Comprehension

54.Vision Transformers

• Dexterous Grasp Transformer
• Mean-Shift Feature Transformer
• MLP Can Be A Good Transformer Learner
  :star:code
• Unifying Top-down and Bottom-up Scanpath Prediction Using Transformers
• Solving Masked Jigsaw Puzzles with Diffusion Vision Transformers
• Dual-scale Transformer for Large-scale Single-Pixel Imaging
• DeiT-LT: Distillation Strikes Back for Vision Transformer Training on Long-Tailed Datasets
• Solving Masked Jigsaw Puzzles with Diffusion Transformers
• Towards Understanding and Improving Adversarial Robustness of Vision Transformers
• RMT: Retentive Networks Meet Vision Transformers
  :star:code
• You Only Need Less Attention at Each Stage in Vision Transformers
• MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers
  :house:project
• Instance-Aware Group Quantization for Vision Transformers
  :house:project
• Multi-criteria Token Fusion with One-step-ahead Attention for Efficient Vision Transformers
  :star:code
• RepViT: Revisiting Mobile CNN From ViT Perspective
  :star:code
• Token Transformation Matters: Towards Faithful Post-hoc Explanation for Vision Transformer
• Autoregressive Queries for Adaptive Tracking with Spatio-Temporal Transformers
  :thumbsup:摘要
• Comparing the Decision-Making Mechanisms by Transformers and CNNs via Explanation Methods
• On the Faithfulness of Vision Transformer Explanations
• Learning Correlation Structures for Vision Transformers
• Low-Rank Rescaled Vision Transformer Fine-Tuning: A Residual Design Approach
  :star:code
• Once for Both: Single Stage of Importance and Sparsity Search for Vision Transformer Compression
• Point Transformer V3: Simpler Faster Stronger
  :star:code
• A General and Efficient Training for Transformer via Token Expansion
  :star:code
• HEAL-SWIN: A Vision Transformer On The Sphere
  :star:code
• SHViT: Single-Head Vision Transformer with Memory Efficient Macro DesignVision
• TransNeXt: Robust Foveal Visual Perception for Vision Transformers
  :star:code
• Making Vision Transformers Truly Shift-Equivariant
• Multimodal Pathway: Improve Transformers with Irrelevant Data from Other Modalities
  :star:code
• Random Entangled Tokens for Adversarially Robust Vision Transformer

53.Light-Field(光场)

• Time-Efficient Light-Field Acquisition Using Coded Aperture and Events
  :house:project
• Continuous Pose for Monocular Cameras in Neural Implicit Representation
  :star:code
• PanoPose: Self-supervised Relative Pose Estimation for Panoramic Images
  :house:project
• Unbiased Estimator for Distorted Conics in Camera Calibration
• 相机姿态
  • Matching 2D Images in 3D: Metric Relative Pose from Metric Correspondences
  • Map-Relative Pose Regression for Visual Re-Localization
    :star:code
  • The Unreasonable Effectiveness of Pre-Trained Features for Camera Pose Refinement
    :star:code
• 快照压缩成像
  • DiffSCI: Zero-Shot Snapshot Compressive Imaging via Iterative Spectral Diffusion Model

52.Biomedical

• ManiFPT: Defining and Analyzing Fingerprints of Generative Models
• Flexible Biometrics Recognition: Bridging the Multimodality Gap through Attention Alignment and Prompt Tuning生物识别
• 人员识别
  • Activity-Biometrics: Person Identification from Daily Activities
    :star:code

51.Style Transfer(风格迁移)

• Z*: Zero-shot Style Transfer via Attention Reweighting
• MoST: Motion Style Transformer Between Diverse Action Contents
  :star:code
• ArtAdapter: Text-to-Image Style Transfer using Multi-Level Style Encoder and Explicit Adaptation
  :star:code
  :house:project
• Arbitrary Motion Style Transfer with Multi-condition Motion Latent Diffusion Model
• Style Injection in Diffusion: A Training-free Approach for Adapting Large-scale Diffusion Models for Style Transfer
  :house:project
• Puff-Net: Efficient Style Transfer with Pure Content and Style Feature Fusion Network
  :thumbsup:平衡效率与质量，南航提出新风格迁移算法Puff-Net
• 零样本文本驱动运动迁移
  • Space-Time Diffusion Features for Zero-Shot Text-Driven Motion Transfer
    :house:project

50.OOD Detection

• Test-Time Linear Out-of-Distribution Detection
• Segment Every Out-of-Distribution Object
• Label-Efficient Group Robustness via Out-of-Distribution Concept Curation
• Enhancing the Power of OOD Detection via Sample-Aware Model SelectionOOD
• Discriminability-Driven Channel Selection for Out-of-Distribution Detection
• CORES: Convolutional Response-based Score for Out-of-distribution Detection
• Learning Transferable Negative Prompts for Out-of-Distribution Detection
  :star:code
• A noisy elephant in the room: Is your out-of-distribution detector robust to label noise?
  :star:code
• Improving Out-of-Distribution Generalization in Graphs via Hierarchical Semantic Environments
• A Noisy Elephant in the Room: Is Your Out-of-Distribution Detector Robust to Label Noise?
• 异常检测
  • Hyperbolic Anomaly Detection
  • Universal Novelty Detection through Adaptive Contrastive Learning
  • Looking 3D: Anomaly Detection with 2D-3D Alignment

49.Dataset(数据集)

• 数据集
  • Multiagent Multitraversal Multimodal Self-Driving: Open MARS Dataset
  • 4D-DRESS: A 4D Dataset of Real-World Human Clothing With Semantic Annotations
  • DiLiGenRT: A Photometric Stereo Dataset with Quantified Roughness and Translucency
  • MULAN: A Multi Layer Annotated Dataset for Controllable Text-to-Image Generation
  • LaMPilot: An Open Benchmark Dataset for Autonomous Driving with Language Model Programs
  • 360Loc: A Dataset and Benchmark for Omnidirectional Visual Localization with Cross-device Queries
  • Towards Automatic Power Battery Detection: New Challenge Benchmark Dataset and Baseline
  • MSU-4S - The Michigan State University Four Seasons Dataset
  • DiVa-360: The Dynamic Visual Dataset for Immersive Neural Fields
  • Event Stream-based Visual Object Tracking: A High-Resolution Benchmark Dataset and A Novel Baseline
  • LiDAR-Net: A Real-scanned 3D Point Cloud Dataset for Indoor Scenes
  • Advancing Saliency Ranking with Human Fixations: Dataset, Models and Benchmarks
  • MAGICK: A Large-scale Captioned Dataset from Matting Generated Images using Chroma Keying
  • HardMo: A Large-Scale Hardcase Dataset for Motion Capture
  • The STVchrono Dataset: Towards Continuous Change Recognition in Time
  • Insect-Foundation: A Foundation Model and Large-scale 1M Dataset for Visual Insect Understanding
  • LED: A Large-scale Real-world Paired Dataset for Event Camera Denoising
  • On the Diversity and Realism of Distilled Dataset: An Efficient Dataset Distillation Paradigm
  • Towards Modern Image Manipulation Localization: A Large-Scale Dataset and Novel Methods
  • Habitat Synthetic Scenes Dataset (HSSD-200): An Analysis of 3D Scene Scale and Realism Tradeoffs for ObjectGoal Navigation
  • FineSports: A Multi-person Hierarchical Sports Video Dataset for Fine-grained Action Understanding细粒度动作理解
  • MMSum: A Dataset for Multimodal Summarization and Thumbnail Generation of Videos
    :house:project
  • Traffic Scene Parsing through the TSP6K Dataset
  • Spectral and Polarization Vision: Spectro-polarimetric Real-world Dataset
  • RGBD Objects in the Wild: Scaling Real-World 3D Object Learning from RGB-D Videos
    :house:project
    :sunflower:datasetRGB-D object数据集
  • eTraM: Event-based Traffic Monitoring Dataset
    :star:code
    :house:project流量监控数据集
  • Towards Real-World HDR Video Reconstruction: A Large-Scale Benchmark Dataset and A Two-Stage Alignment Network
    :sunflower:dataset
  • JRDB-Social: A Multifaceted Robotic Dataset for Understanding of Context and Dynamics of Human Interactions Within Social Groups
    :house:project
  • TULIP: Multi-camera 3D Precision Assessment of Parkinson's Disease
  • JRDB-PanoTrack: An Open-world Panoptic Segmentation and Tracking Robotic Dataset in Crowded Human Environments
  • OAKINK2: A Dataset of Bimanual Hands-Object Manipulation in Complex Task Completion
    :house:project
  • SportsHHI: A Dataset for Human-Human Interaction Detection in Sports Videos
  • RELI11D: A Comprehensive Multimodal Human Motion Dataset and Method
    :house:project
    :thumbsup:摘要
  • MatSynth: A Modern PBR Materials Dataset
    :house:project
  • RCooper: A Real-world Large-scale Dataset for Roadside Cooperative Perception
    :star:code
  • Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detection
  • EgoExoLearn: A Dataset for Bridging Asynchronous Ego- and Exo-centric View of Procedural Activities in Real World
    :star:code
  • MCD: Diverse Large-Scale Multi-Campus Dataset for Robot Perception
    :sunflower:dataset
  • HouseCat6D -- A Large-Scale Multi-Modal Category Level 6D Object Perception Dataset with Household Objects in Realistic Scenarios
  • HoloVIC: Large-scale Dataset and Benchmark for Multi-Sensor Holographic Intersection and Vehicle-Infrastructure Cooperative
    :sunflower:dataset
  • DL3DV-10K: A Large-Scale Scene Dataset for Deep Learning-based 3D Vision
    :sunflower:dataset
  • EFHQ: Multi-purpose ExtremePose-Face-HQ dataset
    :star:code
    :house:project数据集
  • LUWA Dataset: Learning Lithic Use-Wear Analysis on Microscopic Images
  • MMVP: A Multimodal MoCap Dataset with Vision and Pressure Sensors
    :star:code
  • FreeMan: Towards Benchmarking 3D Human Pose Estimation under Real-World Conditions
    :house:project
  • TUMTraf V2X Cooperative Perception Dataset
    :house:project
  • MVHumanNet: A Large-scale Dataset of Multi-view Daily Dressing Human Captures
    :sunflower:dataset
• 基准
  • When Visual Grounding Meets Gigapixel-level Large-scale Scenes: Benchmark and Approach
  • THRONE: A Hallucination Benchmark for the Free-form Generations of Large Vision-Language Models
  • M3-UDA: A New Benchmark for Unsupervised Domain Adaptive Fetal Cardiac Structure Detection
    :star:code
  • DriveTrack: A Benchmark for Long-Range Point Tracking in Real-World Videos现实视频中远程点跟踪的基准
  • SOK-Bench: A Situated Video Reasoning Benchmark with Aligned Open-World Knowledge
  • MAPLM: A Real-World Large-Scale Vision-Language Benchmark for Map and Traffic Scene Understanding
  • RoDLA: Benchmarking the Robustness of Document Layout Analysis Models
    :star:code
  • GOAT-Bench: A Benchmark for Multi-Modal Lifelong Navigation
  • MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI
  • Advancing Saliency Ranking with Human Fixations: Dataset Models and Benchmarks
  • ConCon-Chi: Concept-Context Chimera Benchmark for Personalized Vision-Language Tasks
  • Real Acoustic Fields: An Audio-Visual Room Acoustics Dataset and Benchmark
    :star:code
  • UVEB: A Large-scale Benchmark and Baseline Towards Real-World Underwater Video Enhancement
  • PKU-DyMVHumans: A Multi-View Video Benchmark for High-Fidelity Dynamic Human Modeling
    :house:project
  • MVBench: A Comprehensive Multi-modal Video Understanding Benchmark
    :star:code
  • Uncovering What Why and How: A Comprehensive Benchmark for Causation Understanding of Video Anomaly
  • VBench : Comprehensive Benchmark Suite for Video Generative Models
    :star:code
    :house:project
  • MTMMC: A Large-Scale Real-World Multi-Modal Camera Tracking Benchmark
  • CADTalk: An Algorithm and Benchmark for Semantic Commenting of CAD Programs
    :house:project
  • How to Train Neural Field Representations: A Comprehensive Study and Benchmark
  • OmniMedVQA: A New Large-Scale Comprehensive Evaluation Benchmark for Medical LVLM

48.Semi/self-supervised learning(半/自监督)

• 弱监督学习
  • 部分标签学习
    • CroSel: Cross Selection of Confident Pseudo Labels for Partial-Label Learning部分标签学习-弱监督学习问题
• 半监督
  • Targeted Representation Alignment for Open-World Semi-Supervised Learning
  • SeNM-VAE: Semi-Supervised Noise Modeling with Hierarchical Variational Autoencoder
  • CDMAD: Class-Distribution-Mismatch-Aware Debiasing for Class-Imbalanced Semi-Supervised Learning
  • BEM: Balanced and Entropy-based Mix for Long-Tailed Semi-Supervised Learning
  • 正样本标签学习
    • Positive-Unlabeled Learning by Latent Group-Aware Meta DisambiguationPositive-Unlabeled Learning(正样本标签学习)半监督学习的一个重要分支
• 自监督学习
  • Self-supervised Representation Learning from Arbitrary Scenarios
  • Self-supervised Debiasing Using Low Rank Regularization
  • Self-Supervised Dual Contouring
  • Neural Modes: Self-supervised Learning of Nonlinear Modal Subspaces
  • Self-Supervised Representation Learning from Arbitrary Scenarios
  • SD2Event: Self-supervised Learning of Dynamic Detectors and Contextual Descriptors for Event Cameras
  • An Asymmetric Augmented Self-Supervised Learning Method for Unsupervised Fine-Grained Image Hashing
  • Self-supervised debiasing using low rank regularization
  • CNC-Net: Self-Supervised Learning for CNC Machining Operations
• 无监督学习
  • Unsupervised Learning of Category-Level 3D Pose from Object-Centric Videos

47.Dense Predictions(密集预测)

• Efficient Multitask Dense Predictor via Binarization密集预测
• Going Beyond Multi-Task Dense Prediction with Synergy Embedding Models
• Exploiting Diffusion Prior for Generalizable Dense Prediction
  :house:project
• ViT-CoMer: Vision Transformer with Convolutional Multi-scale Feature Interaction for Dense Predictions
  :star:code
  :thumbsup:百度提出视觉新骨干ViT-CoMer，刷新密集预测任务SOTA
• Multi-Task Dense Prediction via Mixture of Low-Rank Experts
  :star:code

46.Industrial Anomaly Detection(工业缺陷检测)

• Anomaly Heterogeneity Learning for Open-set Supervised Anomaly Detection
  :star:code
• 异常检测
  • Supervised Anomaly Detection for Complex Industrial Images
  • Prompt-enhanced Multiple Instance Learning for Weakly Supervised Anomaly Detection弱监督异常检测
  • Multimodal Industrial Anomaly Detection by Crossmodal Feature Mapping
  • Text-Guided Variational Image Generation for Industrial Anomaly Detection and Segmentation
  • Long-Tailed Anomaly Detection with Learnable Class Names
    :house:project
  • RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection
    :star:code
  • Toward Generalist Anomaly Detection via In-context Residual Learning with Few-shot Sample Prompts
    :star:code
  • PromptAD: Learning Prompts with only Normal Samples for Few-Shot Anomaly Detection
    :star:code
• 薄膜去除
  • Learning to Remove Wrinkled Transparent Film with Polarized Prior
    :star:code
• 基准/数据集
  • Real-IAD: A Real-World Multi-view Dataset for Benchmarking Versatile Industrial Anomaly Detection
    :star:code
  • Towards Scalable 3D Anomaly Detection and Localization: A Benchmark via 3D Anomaly Synthesis and A Self-Supervised Learning Network
    :star:code

45.Neural Architecture Search(神经架构搜索)

• Towards Accurate and Robust Architectures via Neural Architecture Search
• Boosting Order-Preserving and Transferability for Neural Architecture Search: a Joint Architecture Refined Search and Fine-tuning Approach
• Building Optimal Neural Architectures using Interpretable Knowledge
  :star:code
• AZ-NAS: Assembling Zero-Cost Proxies for Network Architecture Search
• SNED: Superposition Network Architecture Search for Efficient Video Diffusion Model
• Insights from the Use of Previously Unseen Neural Architecture Search Datasets
• FlowerFormer: Empowering Neural Architecture Encoding using a Flow-aware Graph Transformer
  :star:code

44.Image Fusion(图像融合)

• Equivariant Multi-Modality Image Fusion图像融合
• Task-Customized Mixture of Adapters for General Image Fusion
  :star:code
• Text-IF: Leveraging Semantic Text Guidance for Degradation-Aware and Interactive Image Fusion
  :star:code
• Revisiting Spatial-Frequency Information Integration from a Hierarchical Perspective for Panchromatic and Multi-Spectral Image Fusion
• Neural Spline Fields for Burst Image Fusion and Layer Separation
  :house:project
• 红外和可见光图像融合
  • Probing Synergistic High-Order Interaction in Infrared and Visible Image Fusion

43.Image Matching(图像匹配)

• XFeat: Accelerated Features for Lightweight Image Matching
  :house:project图像匹配
• 图像-文本
  • Composing Object Relations and Attributes for Image-Text Matching

42.Image Retrieval(图像检索)

• Language-only Training of Zero-shot Composed Image Retrieval
  :star:code
• Evaluating Transferability in Retrieval Tasks: An Approach Using MMD and Kernel Methods
• Knowledge-Enhanced Dual-stream Zero-shot Composed Image Retrieval
• On Train-Test Class Overlap and Detection for Image Retrieval
  :star:code
• D3still: Decoupled Differential Distillation for Asymmetric Image Retrieval
• Task-Driven Exploration: Decoupling and Inter-Task Feedback for Joint Moment Retrieval and Highlight Detection
• 跨域检索
  • ProS: Prompting-to-simulate Generalized knowledge for Universal Cross-Domain Retrieval
    :star:code
• 视频检索
  • Composed Video Retrieval via Enriched Context and Discriminative Embeddings
    :star:code
• 跨模态检索
  • Learning to Rematch Mismatched Pairs for Robust Cross-Modal Retrieval
    :star:code
  • Fine-grained Prototypical Voting with Heterogeneous Mixup for Semi-supervised 2D-3D Cross-modal Retrieval
• 文本-视频检索
  • Text Is MASS: Modeling as Stochastic Embedding for Text-Video Retrieval
    :star:code
  • Holistic Features are almost Sufficient for Text-to-Video Retrieval
• 图像-文本检索
  • How to Make Cross Encoder a Good Teacher for Efficient Image-Text Retrieval?
• 视频文本检索
  • MV-Adapter: Multimodal Video Transfer Learning for Video Text Retrieval
• 组合图像检索
  • Visual Delta Generator with Large Multi-modal Models for Semi-supervised Composed Image Retrieval
• 细粒度图像检索
  • You'll Never Walk Alone: A Sketch and Text Duet for Fine-Grained Image Retrieval
    :house:project
  • Characteristics Matching Based Hash Codes Generation for Efficient Fine-grained Image Retrieval
• 基于草图的检索
  • How to Handle Sketch-Abstraction in Sketch-Based Image Retrieval?
    :star:code
  • Text-to-Image Diffusion Models are Great Sketch-Photo Matchmakers
    :house:project

41.Graph Generative Network(GNN/GCN)

• GNN
  • Domain Separation Graph Neural Networks for Saliency Object Ranking
  • GreedyViG: Dynamic Axial Graph Construction for Efficient Vision GNNs
  • FC-GNN: Recovering Reliable and Accurate Correspondences from Interferences
  • DGC-GNN: Leveraging Geometry and Color Cues for Visual Descriptor-Free 2D-3D Matching
  • GLiDR: Topologically Regularized Graph Generative Network for Sparse LiDAR Point Clouds图生成网络
• GCN
  • Learning for Transductive Threshold Calibration in Open-World Recognition

40.Scene Graph Generation(场景图生成)

• Leveraging Predicate and Triplet Learning for Scene Graph Generation
• OED: Towards One-stage End-to-End Dynamic Scene Graph Generation
• CLIP-Driven Open-Vocabulary 3D Scene Graph Generation via Cross-Modality Contrastive Learning
• Multi-Level Neural Scene Graphs for Dynamic Urban Environments
  :star:code
• HiKER-SGG: Hierarchical Knowledge Enhanced Robust Scene Graph Generation
  :star:code
  :star:code
• DSGG: Dense Relation Transformer for an End-to-end Scene Graph Generation
  :star:code
  :house:project
• From Pixels to Graphs: Open-Vocabulary Scene Graph Generation with Vision-Language Models
• EGTR: Extracting Graph from Transformer for Scene Graph Generation
  :star:code
• LLM4SGG: Large Language Models for Weakly Supervised Scene Graph Generation

39.Motion Generation(动作生成)

• Programmable Motion Generation for Open-Set Motion Control Tasks
• Move as You Say Interact as You Can: Language-guided Human Motion Generation with Scene Affordance
• AnySkill: Learning Open-Vocabulary Physical Skill for Interactive Agents
• Towards Variable and Coordinated Holistic Co-Speech Motion Generation
  :star:code
• Generating Human Motion in 3D Scenes from Text Descriptions根据文本描述生成 3D 场景中的人体运动
• NIFTY: Neural Object Interaction Fields for Guided Human Motion Synthesis
  :house:project人体运动合成
• OMG: Towards Open-vocabulary Motion Generation via Mixture of Controllers
  :house:project
• WANDR: Intention-guided Human Motion Generation
  :tv:video
• MAS: Multi-view Ancestral Sampling for 3D Motion Generation Using 2D Diffusion
  :house:project
• Unified-IO 2: Scaling Autoregressive Multimodal Models with Vision Language Audio and Action
• Multimodal Sense-Informed Forecasting of 3D Human Motions
• 运动检索
  • Tri-Modal Motion Retrieval by Learning a Joint Embedding Space
• 动物运动
  • OmniMotionGPT: Animal Motion Generation with Limited Data
    :star:code
    :house:project
• 人体运动预测
  • MoML: Online Meta Adaptation for 3D Human Motion Prediction
  • MoST: Multi-Modality Scene Tokenization for Motion Prediction
  • Rethinking Human Motion Prediction with Symplectic Integral
  • Human Motion Prediction Under Unexpected Perturbation
  • Continual Learning for Motion Prediction Model via Meta-Representation Learning and Optimal Memory Buffer Retention Strategy
• 人体运动估计
  • MultiPhys: Multi-Person Physics-aware 3D Motion Estimation
    :house:project
  • A Unified Diffusion Framework for Scene-aware Human Motion Estimation from Sparse Signals人体运动估计
• 人体运动重建
  • RoHM: Robust Human Motion Reconstruction via Diffusion

38.Vision Question Answering(视觉问答)

• GRAM: Global Reasoning for Multi-Page VQA
• SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities
  :house:project
• Consistency and Uncertainty: Identifying Unreliable Responses From Black-Box Vision-Language Models for Selective Visual Question Answering
• How to Configure Good In-Context Sequence for Visual Question Answering
  :star:code
• Causal-CoG: A Causal-Effect Look at Context Generation for Boosting Multi-modal Language Models
• Question Aware Vision Transformer for Multimodal Reasoning
• OpenEQA: Embodied Question Answering in the Era of Foundation Models
• Video-QA
  • Grounded Question-Answering in Long Egocentric Videos
    :star:code
  • Ranking Distillation for Open-Ended Video Question Answering with Insufficient Labels
  • Language-aware Visual Semantic Distillation for Video Question Answering
  • MoReVQA: Exploring Modular Reasoning Models for Video Question Answering
  • Can I Trust Your Answer? Visually Grounded Video Question Answering
    :star:code
  • Align and Aggregate: Compositional Reasoning with Video Alignment and Answer Aggregation for Video Question-Answering
• 图表问答
  • CoG-DQA: Chain-of-Guiding Learning with Large Language Models for Diagram Question Answering
  • Synthesize Step-by-Step: Tools, Templates and LLMs as Data Generators for Reasoning-Based Chart VQA
• 视觉文本问答
  • VTQA: Visual Text Question Answering via Entity Alignment and Cross-Media Reasoning

37.OCR

• 场景文本识别
  • OTE: Exploring Accurate Scene Text Recognition Using One Token
  • An Empirical Study of Scaling Law for Scene Text Recognition
    :star:code场景文本识别
  • Multi-modal In-Context Learning Makes an Ego-evolving Scene Text Recognizer
    :star:code
  • Kernel Adaptive Convolution for Scene Text Detection via Distance Map Prediction
  • Choose What You Need: Disentangled Representation Learning for Scene Text Recognition Removal and Editing场景文本识别、删除和编辑
  • ODM: A Text-Image Further Alignment Pre-training Approach for Scene Text Detection and Spotting
    :star:code
• 场景文本图像合成
  • Layout-Agnostic Scene Text Image Synthesis with Diffusion Models
• 场景文本理解
  • LayoutFormer: Hierarchical Text Detection Towards Scene Text Understanding
• 化学结构识别
  • Atom-Level Optical Chemical Structure Recognition with Limited Supervision
    :star:code
• 文档色度检测
  • CMA: A Chromaticity Map Adapter for Robust Detection of Screen-Recapture Document Images
    :star:code
• 文本检测
  • OmniParser: A Unified Framework for Text Spotting Key Information Extraction and Table Recognition
    :star:code
  • Bridging the Gap Between End-to-End and Two-Step Text Spotting
    :star:code
  • Text Grouping Adapter: Adapting Pre-trained Text Detector for Layout Analysis
• 文档理解
  • LayoutLLM: Layout Instruction Tuning with Large Language Models for Document Understanding
    :star:code
  • HRVDA: High-Resolution Visual Document Assistant
• 字体生成
  • Generate Like Experts: Multi-Stage Font Generation by Incorporating Font Transfer Process into Diffusion Models

36.4D Reconstruction(4D 重建)

• Gaussian-Flow: 4D Reconstruction with Dynamic 3D Gaussian Particle
  :house:project
• Motion2VecSets: 4D Latent Vector Set Diffusion for Non-rigid Shape Reconstruction and Tracking
  :house:project
• 4D Gaussian Splatting for Real-Time Dynamic Scene Rendering
  :star:code
  :house:project
• 文本和图像引导 4D 场景生成
  • A Unified Approach for Text- and Image-guided 4D Scene Generation
    :house:project
• 4D视图合成
  • 4K4D: Real-Time 4D View Synthesis at 4K Resolution
    :star:code
    :house:project
• 语言到 4D 建模
  • L4D-Track: Language-to-4D Modeling Towards 6-DoF Tracking and Shape Reconstruction in 3D Point Cloud Stream
    :star:code

35.Scene Understanding(场景理解)

• Omni-Q: Omni-Directional Scene Understanding for Unsupervised Visual Grounding
• PanoContext-Former: Panoramic Total Scene Understanding with a Transformer
• DriveWorld: 4D Pre-trained Scene Understanding via World Models for Autonomous Driving
• OpenESS: Event-based Semantic Scene Understanding with Open Vocabularies
  :star:code
• A Category Agnostic Model for Visual Rearrangment
  :thumbsup:VILP
• 360+x: A Panoptic Multi-modal Scene Understanding Dataset
  :star:code
• 开放词汇场景理解
  • Language Embedded 3D Gaussians for Open-Vocabulary Scene Understanding
• 3D场景理解
  • HUGS: Holistic Urban 3D Scene Understanding via Gaussian Splatting
    :house:project
  • SurroundSDF: Implicit 3D Scene Understanding Based on Signed Distance Field
  • GSNeRF: Generalizable Semantic Neural Radiance Fields with Enhanced 3D Scene Understanding
  • GP-NeRF: Generalized Perception NeRF for Context-Aware 3D Scene Understanding
  • RegionPLC: Regional Point-Language Contrastive Learning for Open-World 3D Scene Understanding
    :house:project
  • GroupContrast: Semantic-aware Self-supervised Representation Learning for 3D Understanding
    :star:code
  • SceneFun3D: Fine-Grained Functionality and Affordance Understanding in 3D Scenes

34.Human–Computer Interaction(人机交互)

• Exploring Pose-Aware Human-Object Interaction via Hybrid Learning
• Bilateral Adaptation for Human-Object Interaction Detection with Occlusion-Robustness
• Scaling Up Dynamic Human-Scene Interaction Modeling
  :star:code
  :house:project
• ReGenNet: Towards Human Action-Reaction Synthesis
  :star:code
• DRESS: Instructing Large Vision-Language Models to Align and Interact with Humans via Natural Language Feedback
  :star:code交互
• HOI-M^3: Capture Multiple Humans and Objects Interaction within Contextual Environment
• GenZI: Zero-Shot 3D Human-Scene Interaction Generation
• Exploring the Potential of Large Foundation Models for Open-Vocabulary HOI Detection
• 人体运动跟踪
  • HMD-Poser: On-Device Real-time Human Motion Tracking from Scalable Sparse Observations
    :star:code
    :house:project
• 新运动合成
  • PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics
    :star:code
    :house:project
• 手部交互
  • InterHandGen: Two-Hand Interaction Generation via Cascaded Reverse Diffusion
    :star:code
  • HOIDiffusion: Generating Realistic 3D Hand-Object Interaction Data
  • Physics-Aware Hand-Object Interaction Denoising
  • HOLD: Category-agnostic 3D Reconstruction of Interacting Hands and Objects from Video
    :star:code
    :house:project手物交互
  • GEARS: Local Geometry-aware Hand-object Interaction Synthesis
  • TACO: Benchmarking Generalizable Bimanual Tool-ACtion-Object Understanding
    :house:project
  • Text2HOI: Text-guided 3D Motion Generation for Hand-Object Interaction
    :star:code
  • G-HOP: Generative Hand-Object Prior for Interaction Reconstruction and Grasp Synthesis
    :star:code
  • MOHO: Learning Single-view Hand-held Object Reconstruction with Multi-view Occlusion-Aware Supervision
  • HOIST-Former: Hand-held Objects Identification Segmentation and Tracking in the Wild
• 人物交互
  • Discovering Syntactic Interaction Clues for Human-Object Interaction Detection
  • Open-World Human-Object Interaction Detection via Multi-modal Prompts
  • LEMON: Learning 3D Human-Object Interaction Relation from 2D Images
    :star:code
    :house:project
  • Disentangled Pre-training for Human-Object Interaction Detection
    :star:code
  • GenH2R: Learning Generalizable Human-to-Robot Handover via Scalable Simulation Demonstration and Imitation
    :house:project
  • Learning from Observer Gaze: Zero-Shot Attention Prediction Oriented by Human-Object Interaction Recognition
    :house:project
  • Template Free Reconstruction of Human-object Interaction with Procedural Interaction Generation
    :house:project
  • 3D 人物交互
    • I'M HOI: Inertia-aware Monocular Capture of 3D Human-Object Interactions
      :house:project
    • CG-HOI: Contact-Guided 3D Human-Object Interaction Generation
      :house:project
• 人-人交互
  • Inter-X: Towards Versatile Human-Human Interaction Analysis
    :star:code
    :house:project
    :thumbsup:三维数字人重建、编辑与驱动

33.NeRF

• GARField: Group Anything with Radiance Fields
• IReNe: Instant Recoloring of Neural Radiance Fields
• PIE-NeRF: Physics-based Interactive Elastodynamics with NeRF
• LidaRF: Delving into Lidar for Neural Radiance Field on Street Scenes
• SIGNeRF: Scene Integrated Generation for Neural Radiance Fields
• NC-SDF: Enhancing Indoor Scene Reconstruction Using Neural SDFs with View-Dependent Normal Compensation
• SpecNeRF: Gaussian Directional Encoding for Specular Reflections
• PaReNeRF: Toward Fast Large-scale Dynamic NeRF with Patch-based ReferenceNeRF
• Global and Hierarchical Geometry Consistency Priors for Few-shot NeRFs in Indoor Scenes
  :thumbsup:摘要
• NeRF Analogies: Example-Based Visual Attribute Transfer for NeRFs
• Neural Directional Encoding for Efficient and Accurate View-Dependent Appearance Modeling
  :star:code
• Accelerating Neural Field Training via Soft Mining
• Gear-NeRF: Free-Viewpoint Rendering and Tracking with Motion-aware Spatio-Temporal Sampling
  :house:project
• How Far Can We Compress Instant-NGP-Based NeRF?
  :star:code
  :house:project
• BANF: Band-Limited Neural Fields for Levels of Detail Reconstruction
  :star:code
  :house:project
• Tactile-Augmented Radiance Fields
  :star:code
  :house:project
• NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild
  :house:project
• L0-Sampler: An L0 Model Guided Volume Sampling for NeRF
  :house:projectNeRF
• HumanNeRF-SE: A Simple yet Effective Approach to Animate HumanNeRF with Diverse Poses
• Entangled View-Epipolar Information Aggregation for Generalizable Neural Radiance Fields
  :star:code
• NeRFCodec: Neural Feature Compression Meets Neural Radiance Fields for Memory-Efficient Scene Representation
• MuRF: Multi-Baseline Radiance Fields
  :house:project
  :house:project
• InNeRF360: Text-Guided 3D-Consistent Object Inpainting on 360-degree Neural Radiance Fields
• NRDF: Neural Riemannian Distance Fields for Learning Articulated Pose Priors
  :star:code
  :house:project
• Neural Fields as Distributions: Signal Processing Beyond Euclidean Space
  :house:project
• CVT-xRF: Contrastive In-Voxel Transformer for 3D Consistent Radiance Fields from Sparse Inputs
  :star:code
• DaReNeRF: Direction-aware Representation for Dynamic Scenes
• Geometry Transfer for Stylizing Radiance Fields
  :house:project
• S-DyRF: Reference-Based Stylized Radiance Fields for Dynamic Scenes
  :star:code
• SpikeNeRF: Learning Neural Radiance Fields from Continuous Spike Stream
  :star:code
• Entity-NeRF: Detecting and Removing Moving Entities in Urban Scenes
  :star:code
• Language-driven Object Fusion into Neural Radiance Fields with Pose-Conditioned Dataset Updates
  :star:code
• LAENeRF: Local Appearance Editing for Neural Radiance Fields
  :star:code
  :house:project
• Single View Refractive Index Tomography with Neural Fields
• ExtraNeRF: Visibility-Aware View Extrapolation of Neural Radiance Fields with Diffusion Models
• TeTriRF: Temporal Tri-Plane Radiance Fields for Efficient Free-Viewpoint Video
• NeRF-HuGS: Improved Neural Radiance Fields in Non-static Scenes Using Heuristics-Guided Segmentation
  :star:code
• Learning with Unreliability: Fast Few-shot Voxel Radiance Fields with Relative Geometric Consistency
  :star:code
• Grounding and Enhancing Grid-based Models for Neural Fields
  :house:project
• Mitigating Motion Blur in Neural Radiance Fields with Events and Frames
• OmniLocalRF: Omnidirectional Local Radiance Fields from Dynamic Videos
• Neural Implicit Representation for Building Digital Twins of Unknown Articulated Objects
  :star:code
• Bayes' Rays: Uncertainty Quantification for Neural Radiance Fields
• Alpha Invariance: On Inverse Scaling Between Distance and Volume Density in Neural Radiance Fields
  :house:project
• Dynamic LiDAR Re-simulation using Compositional Neural Fields
  :house:project
• SOAC: Spatio-Temporal Overlap-Aware Multi-Sensor Calibration using Neural Radiance Fields
  :house:project
• ICON: Incremental CONfidence for Joint Pose and Radiance Field Optimization
• NeRFDeformer: NeRF Transformation from a Single View via 3D Scene Flows
• 新视图合成
  • ZeroNVS: Zero-Shot 360-Degree View Synthesis from a Single Image
  • Unifying Correspondence Pose and NeRF for Generalized Pose-Free Novel View Synthesis
  • NeLF-Pro: Neural Light Field Probes for Multi-Scale Novel View Synthesis
  • 3D Geometry-Aware Deformable Gaussian Splatting for Dynamic View Synthesis
  • G-NeRF: Geometry-enhanced Novel View Synthesis from Single-View Images
  • MultiDiff: Consistent Novel View Synthesis from a Single Image
  • Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesis
  • DiffPortrait3D: Controllable Diffusion for Zero-Shot Portrait View Synthesis
  • 3D Geometry-aware Deformable Gaussian Splatting for Dynamic View Synthesis
  • Generalizable Novel-View Synthesis using a Stereo Camera
    :house:project
  • DART: Implicit Doppler Tomography for Radar Novel View Synthesis
    :house:project
  • XScale-NVS: Cross-Scale Novel View Synthesis with Hash Featurized Manifold
  • Spacetime Gaussian Feature Splatting for Real-Time Dynamic View Synthesis
    :star:code
    :house:project
  • NViST: In the Wild New View Synthesis from a Single Image with Transformers
    :star:code
    :house:project
  • ViVid-1-to-3: Novel View Synthesis with Video Diffusion Models
    :house:project
  • SC-GS: Sparse-Controlled Gaussian Splatting for Editable Dynamic Scenes
    :star:code
    :house:project
  • Neural Visibility Field for Uncertainty-Driven Active Mapping
    :house:project
  • EscherNet: A Generative Model for Scalable View Synthesis
    :star:code
    :house:project
  • GPS-Gaussian: Generalizable Pixel-wise 3D Gaussian Splatting for Real-time Human Novel View Synthesis
    :star:code
    :house:project新视图
  • DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalization
    :star:code
    :house:project
  • LiDAR4D: Dynamic Neural Fields for Novel Space-time View LiDAR Synthesis
    :star:code
    :house:project
  • Is Vanilla MLP in Neural Radiance Field Enough for Few-shot View Synthesis?
  • Q-Instruct: Improving Low-level Visual Abilities for Multi-modality Foundation Models
    :star:code
    :house:project
  • CoPoNeRF: Unifying Correspondence, Pose and NeRF for Pose-Free Novel View Synthesis from Stereo Pairs
    :star:code
    :house:project
  • EpiDiff: Enhancing Multi-View Synthesis via Localized Epipolar-Constrained Diffusion
    :star:code
    :house:project
  • Free3D: Consistent Novel View Synthesis without 3D Representation
    :star:code
    :house:project
  • Novel View Synthesis with View-Dependent Effects from a Single Image
    :house:project
• 渲染
  • NeRF Director: Revisiting View Selection in Neural Volume Rendering
  • Multiplane Prior Guided Few-Shot Aerial Scene Rendering渲染
  • Differentiable Point-based Inverse Rendering逆渲染
  • Diffusion Reflectance Map: Single-Image Stochastic Inverse Rendering of Illumination and Reflectance渲染
  • Perceptual Assessment and Optimization of HDR Image Rendering
  • Global Latent Neural Rendering
  • Geometry-aware Reconstruction and Fusion-refined Rendering for Generalizable Neural Radiance Fields
    :star:code
    :house:project
  • GES: Generalized Exponential Splatting for Efficient Radiance Field Rendering
    :house:project
  • Real-time Acquisition and Reconstruction of Dynamic Volumes with Neural Structured Illumination
    :house:project
    :tv:video
    :thumbsup:借助神经结构光，浙大实现动态三维现象的实时采集重建
  • Inverse Rendering of Glossy Objects via the Neural Plenoptic Function and Radiance Fields
    :house:project
  • Dr.Bokeh: DiffeRentiable Occlusion-aware Bokeh Rendering
    :house:project
  • HiFi4G: High-Fidelity Human Performance Rendering via Compact Gaussian Splatting
    :thumbsup:HiFi4G: 通过紧凑高斯进行高保真人体性能渲染
  • ASH: Animatable Gaussian Splats for Efficient and Photoreal Human Rendering
    :house:project
  • SHINOBI: Shape and Illumination using Neural Object Decomposition via BRDF Optimization In-the-wild
    :house:project神经渲染
  • LTM: Lightweight Textured Mesh Extraction and Refinement of Large Unbounded Scenes for Efficient Storage and Real-time Rendering
  • HashPoint: Accelerated Point Searching and Sampling for Neural Rendering
    :house:project
  • HybridNeRF: Efficient Neural Rendering via Adaptive Volumetric Surfaces
    :house:project
  • DUDF: Differentiable Unsigned Distance Fields with Hyperbolic Scaling
    :star:code
    :house:project
  • Holoported Characters: Real-time Free-viewpoint Rendering of Humans from Sparse RGB Cameras
    :house:project
  • ConTex-Human: Free-View Rendering of Human from a Single Image with Texture-Consistent Synthesis
    :house:project
• 多视图逆渲染
  • VMINer: Versatile Multi-view Inverse Rendering with Near- and Far-field Light Sources
• 目标重建
  • Neural Parametric Gaussians for Monocular Non-Rigid Object Reconstruction
    :house:project
  • SAOR: Single-View Articulated Object Reconstruction
    :house:project

32.NLP(自然语言处理)

• Describing Differences in Image Sets with Natural Language
• 实体识别
  • A Generative Approach for Wikipedia-Scale Visual Entity Recognition
• 提示学习
  • BadCLIP: Trigger-Aware Prompt Learning for Backdoor Attacks on CLIP
  • Active Prompt Learning in Vision Language Models
    :star:code
  • Domain Prompt Learning with Quaternion Networks
  • On the Test-Time Zero-Shot Generalization of Vision-Language Models: Do We Really Need Prompt Learning?
  • ID-like Prompt Learning for Few-Shot Out-of-Distribution Detection
• 基础模型
  • Asymmetric Masked Distillation for Pre-Training Small Foundation Models
    :star:code
  • Bootstrapping SparseFormers from Vision Foundation Models
    :star:code

31.Edge Detection(边缘检测)

• MuGE: Multiple Granularity Edge Detection
• RankED: Addressing Imbalance and Uncertainty in Edge Detection Using Ranking-based Losses
  :star:code

30.Person Re-Identification(人员重识别)

• Fusing Personal and Environmental Cues for Identification and Segmentation of First-Person Camera Wearers in Third-Person Views
• Evidential Active Recognition: Intelligent and Prudent Open-World Embodied Perception
• 行人检测
  • DAP: A Dynamic Adversarial Patch for Evading Person Detectors
  • Causal Mode Multiplexer: A Novel Framework for Unbiased Multispectral Pedestrian Detection
    :star:code
  • WALT3D: Generating Realistic Training Data from Time-Lapse Imagery for Reconstructing Dynamic Objects Under Occlusion
  • 基于文本的行人检索
    • UFineBench: Towards Text-based Person Retrieval with Ultra-fine Granularity
      :star:code
• 人群计数
  • Single Domain Generalization for Crowd Counting
    :star:code
  • CrowdDiff: Multi-hypothesis Crowd Density Estimation using Diffusion Models
    :house:project
  • Regressor-Segmenter Mutual Prompt Learning for Crowd Counting
• 行人属性检测
  • Learning Group Activity Features Through Person Attribute Prediction
    :star:code
    :house:project
• 重识别
  • SEAS: ShapE-Aligned Supervision for Person Re-Identification
  • Learning Continual Compatible Representation for Re-indexing Free Lifelong Person Re-identification
    :star:code
  • View-decoupled Transformer for Person Re-identification under Aerial-ground Camera Network
    :star:code
  • CA-Jaccard: Camera-aware Jaccard Distance for Person Re-identification
  • Attribute-Guided Pedestrian Retrieval: Bridging Person Re-ID with Internal Attribute Variability
  • All in One Framework for Multimodal Re-identification in the Wild
  • A Pedestrian is Worth One Prompt: Towards Language Guidance Person Re-Identification
  • Distribution-aware Knowledge Prototyping for Non-exemplar Lifelong Person Re-identification
  • Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions
    :star:code
  • 基于雷达的Re-Id
    • LiDAR-based Person Re-identification
  • 可见光-红外人员重识别
    • Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification
      :star:code
    • Shallow-Deep Collaborative Learning for Unsupervised Visible-Infrared Person Re-Identification
  • 文本-图像重识别
    • Harnessing the Power of MLLMs for Transferable Text-to-Image Person ReID
    • Noisy-Correspondence Learning for Text-to-Image Person Re-identification
      :star:code
• 步态识别
  • Learning Visual Prompt for Gait Recognition
  • BigGait: Learning Gait Representation You Want by Large Vision Models
    :star:code

29.Model Compression/Knowledge Distillation/Pruning(模型压缩/知识蒸馏/剪枝)

• MC
  • Dense Vision Transformer Compression with Few Samples
• KD
  • Small Scale Data-Free Knowledge Distillation
  • KD-DETR: Knowledge Distillation for Detection Transformer with Consistent Distillation Points Sampling
  • Boosting Self-Supervision for Single-View Scene Completion via Knowledge Distillation
  • Correlation-Decoupled Knowledge Distillation for Multimodal Sentiment Analysis with Incomplete Modalities
  • C2KD: Bridging the Modality Gap for Cross-Modal Knowledge Distillation
  • CrossKD: Cross-Head Knowledge Distillation for Object Detection
  • CLIP-KD: An Empirical Study of CLIP Model Distillation
    :star:code
  • Aligning Logits Generatively for Principled Black-Box Knowledge Distillation
  • FreeKD: Knowledge Distillation via Semantic Frequency Prompt
  • Logit Standardization in Knowledge Distillation
  • $V_kD:$ Improving Knowledge Distillation using Orthogonal Projections
    :star:code
  • Scale Decoupled Distillation
    :star:code
  • NAYER: Noisy Layer Data Generation for Efficient and Effective Data-free Knowledge Distillation
    :star:code
  • De-confounded Data-free Knowledge Distillation for Handling Distribution Shifts
  • PromptKD: Unsupervised Prompt Distillation for Vision-Language Models
    :star:code
    :house:project
    :thumbsup:中文解读
• 剪枝
  • Device-Wise Federated Network Pruning
  • FedMef: Towards Memory-efficient Federated Dynamic Pruning
  • OrthCaps: An Orthogonal CapsNet with Sparse Attention Routing and Pruning
  • BilevelPruning: Unified Dynamic and Static Channel Pruning for Convolutional Neural Networks
  • Resource-Efficient Transformer Pruning for Finetuning of Large Models
  • Auto-Train-Once: Controller Network Guided Automatic Network Pruning from Scratch
  • Finding Lottery Tickets in Vision Models via Data-driven Spectral Foresight Pruning
    :house:project
  • Zero-TPrune: Zero-Shot Token Pruning through Leveraging of the Attention Graph in Pre-Trained Transformers
  • MoPE-CLIP: Structured Pruning for Efficient Vision-Language Models with Module-wise Pruning Error Metric
  • Jointly Training and Pruning CNNs via Learnable Agent Guidance and Alignment
  • MULTIFLOW: Shifting Towards Task-Agnostic Vision-Language Pruning
    :star:code
• 量化
  • PTQ4SAM: Post-Training Quantization for Segment Anything
  • Reg-PTQ: Regression-specialized Post-training Quantization for Fully Quantized Object Detector
  • Data-Free Quantization via Pseudo-label Filtering
  • JointSQ: Joint Sparsification-Quantization for Distributed Learning
  • Retraining-Free Model Quantization via One-Shot Weight-Coupling Learning
  • Epistemic Uncertainty Quantification For Pre-Trained Neural Networks
  • Enhancing Post-training Quantization Calibration through Contrastive Learning
  • Towards Accurate Post-training Quantization for Diffusion Models量化
  • Is Conventional SNN Really Efficient? A Perspective from Network Quantization
  • Are Conventional SNNs Really Efficient? A Perspective from Network Quantization

28.UAV/Remote Sensing/Satellite Image(无人机/遥感/卫星图像)

• Unleashing Unlabeled Data: A Paradigm for Cross-View Geo-Localization
  :star:code
• Rethinking Transformers Pre-training for Multi-Spectral Satellite Imagery
  :star:code
• Aerial Lifting: Neural Urban Semantic and Building Instance Lifting from Aerial Imagery
  :house:project
• S2MAE: A Spatial-Spectral Pretraining Foundation Model for Spectral Remote Sensing Data
• Learnable Earth Parser: Discovering 3D Prototypes in Aerial Scans
  :house:project
• WildlifeMapper: Aerial Image Analysis for Multi-Species Detection and Identification
  :star:code
• Learning without Exact Guidance: Updating Large-scale High-resolution Land Cover Maps from Low-resolution Historical Labels
  :star:code
• 遥感
  • GeoChat: Grounded Large Vision-Language Model for Remote Sensing
  • SkySense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagery
  • 3D Building Reconstruction from Monocular Remote Sensing Images with Multi-level Supervisions
    :star:code
  • Poly Kernel Inception Network for Remote Sensing Detection
  • Content-Adaptive Non-Local Convolution for Remote Sensing Pansharpening
    :star:code
• 航空图像分割
  • SatSynth: Augmenting Image-Mask Pairs through Diffusion Models for Aerial Semantic Segmentation
  • Rotated Multi-Scale Interaction Network for Referring Remote Sensing Image Segmentation
    :star:code
    :thumbsup:摘要
• 基于参考图像的超分辨率
  • Building Bridges across Spatial and Temporal Resolutions: Reference-Based Super-Resolution via Change Priors and Conditional Diffusion Model
    :star:code
• 基于UAV的目标检测
  • Weakly Misalignment-free Adaptive Feature Alignment for UAVs-based Multimodal Object Detection
• 交叉视角定位
  • View From Above: Orthogonal-View aware Cross-view Localization

27.Vision-Language(视觉语言)

• A Vision Check-up for Language Models
• The Neglected Tails in Vision-Language Models
• Beyond Average: Individualized Visual Scanpath Prediction
• ArGue: Attribute-Guided Prompt Tuning for Vision-Language Models
• Language Models as Black-Box Optimizers for Vision-Language Models
• Distilling Vision-Language Models on Millions of Videos
• SonicVisionLM: Playing Sound with Vision Language Models
• Jack of All Tasks Master of Many: Designing General-Purpose Coarse-to-Fine Vision-Language Model
• Visual Program Distillation: Distilling Tools and Programmatic Reasoning into Vision-Language Models
• JoAPR: Cleaning the Lens of Prompt Learning for Vision-Language Models
• MMA: Multi-Modal Adapter for Vision-Language Models
• Linguistic-Aware Patch Slimming Framework for Fine-grained Cross-Modal Alignment
• Building Vision-Language Models on Solid Foundations with Masked Distillation
• TCP:Textual-based Class-aware Prompt tuning for Visual-Language Model
  :star:code
• On Scaling Up a Multilingual Vision and Language Model
• CogAgent: A Visual Language Model for GUI Agents
  :star:code
• Towards Better Vision-Inspired Vision-Language Models
• SaCo Loss: Sample-wise Affinity Consistency for Vision-Language Pre-training
• MADTP: Multimodal Alignment-Guided Dynamic Token Pruning for Accelerating Vision-Language Transformer
• Sequential Modeling Enables Scalable Learning for Large Vision Models
  :house:project大型视觉模型
• Seeing the Unseen: Visual Common Sense for Semantic Placement
• Efficient Vision-Language Pre-training by Cluster Masking
  :star:code
  :house:project
• VILA: On Pre-training for Visual Language Models
• EgoThink: Evaluating First-Person Perspective Thinking Capability of Vision-Language Models
  :star:code
  :house:project
• SPIN: Simultaneous Perception Interaction and Navigation
• MAFA: Managing False Negatives for Vision-Language Pre-training
• Visual In-Context Prompting
  :star:code
• Semantics-aware Motion Retargeting with Vision-Language Models
• DePT: Decoupled Prompt Tuning
  :star:code
• Osprey: Pixel Understanding with Visual Instruction Tuning
  :star:code
• FairCLIP: Harnessing Fairness in Vision-Language Learning
  :house:project
• Efficient Test-Time Adaptation of Vision-Language Models
  :star:code
• BioCLIP: A Vision Foundation Model for the Tree of Life
  :star:code
• InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks
  :star:code
• Anchor-based Robust Finetuning of Vision-Language Models
• Multi-Modal Hallucination Control by Visual Information Grounding
• Do Vision and Language Encoders Represent the World Similarly?
• Dual-View Visual Contextualization for Web Navigation
• Any-Shift Prompting for Generalization over Distributions
• Non-autoregressive Sequence-to-Sequence Vision-Language Models
• One Prompt Word is Enough to Boost Adversarial Robustness for Pre-trained Vision-Language Models
  :star:code
• SC-Tune: Unleashing Self-Consistent Referential Comprehension in Large Vision Language Models
  :star:code
• RegionGPT: Towards Region Understanding Vision Language Model
• Enhancing Vision-Language Pre-training with Rich Supervisions
• Grounding Everything: Emerging Localization Properties in Vision-Language Transformers
  :star:code
• Probing and Mitigating Intersectional Social Biases in Vision-Language Models with Counterfactual Examples
• Beyond Text: Frozen Large Language Models in Visual Signal Comprehension
  :star:code
• Contrasting Intra-Modal and Ranking Cross-Modal Hard Negatives to Enhance Visio-Linguistic Compositional Understanding
  :star:code视觉语言构图理解
• FFF: Fixing Flawed Foundations in Contrastive Pre-Training Results in Very Strong Vision-Language Models
• [Enhancing Vision-Language Pretraining with Rich Supervisions]
• Improved Baselines with Visual Instruction Tuning
  :house:project
• Semantic Shield: Defending Vision-Language Models Against Backdooring and Poisoning via Fine-grained Knowledge Alignment
• Dual Memory Networks: A Versatile Adaptation Approach for Vision-Language Models
  :star:code
• A Closer Look at the Few-Shot Adaptation of Large Vision-Language Models
  :star:code
• Synthesize, Diagnose, and Optimize: Towards Fine-Grained Vision-Language Understanding
  :star:code
• SyncMask: Synchronized Attentional Masking for Fashion-centric Vision-Language Pretraining视觉-语言
• Learning by Correction: Efficient Tuning Task for Zero-Shot Generative Vision-Language Reasoning
• Improving Visual Recognition with Hyperbolical Visual Hierarchy Mapping
  :star:code
• Iterated Learning Improves Compositionality in Large Vision-Language Models
• ViTamin: Designing Scalable Vision Models in the Vision-Language Era
  :star:code
• Pre-trained Vision and Language Transformers Are Few-Shot Incremental Learners
  :star:code
• Visual Concept Connectome (VCC): Open World Concept Discovery and their Interlayer Connections in Deep Models
  :house:project
• Know Your Neighbors: Improving Single-View Reconstruction via Spatial Vision-Language Reasoning
  :house:project
• HallusionBench: An Advanced Diagnostic Suite for Entangled Language Hallucination and Visual Illusion in Large Vision-Language Models
  :star:code
• Enhancing Visual Document Understanding with Contrastive Learning in Large Visual-Language Models
• Learning Vision from Models Rivals Learning Vision from Data
  :star:code
• Probing the 3D Awareness of Visual Foundation Models
  :star:code
• LL3DA: Visual Interactive Instruction Tuning for Omni-3D Understanding Reasoning and Planning
  :house:project
• 视觉理解
  • Towards More Unified In-context Visual Understanding
• LLM
  • PixelLM: Pixel Reasoning with Large Multimodal Model
    :house:project
  • OneLLM: One Framework to Align All Modalities with Language
  • Embodied Multi-Modal Agent trained by an LLM from a Parallel TextWorld
  • Let's Think Outside the Box: Exploring Leap-of-Thought in Large Language Models with Creative Humor Generation
  • Learning to Localize Objects Improves Spatial Reasoning in Visual-LLMs
  • Hallucination Augmented Contrastive Learning for Multimodal Large Language Model
  • See Say and Segment: Teaching LMMs to Overcome False Premises
  • ViP-LLaVA: Making Large Multimodal Models Understand Arbitrary Visual Prompts
  • Driving Everywhere with Large Language Model Policy Adaptation
    :house:project
  • Exploring the Transferability of Visual Prompting for Multimodal Large Language Models
  • GROUNDHOG: Grounding Large Language Models to Holistic Segmentation
    :house:project
  • Self-Training Large Language Models for Improved Visual Program Synthesis With Visual Reinforcement
    :house:project
  • V*: Guided Visual Search as a Core Mechanism in Multimodal LLMs
    :star:code
  • Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding
  • Pixel Aligned Language Models
    :house:project
  • SNIFFER: Multimodal Large Language Model for Explainable Out-of-Context Misinformation Detection
  • OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocation
    :star:code多模态大语言模型
  • Low-Rank Approximation for Sparse Attention in Multi-Modal LLMsLLMs
  • LISA: Reasoning Segmentation via Large Language Model
    :star:code
  • Querying as Prompt: Parameter-Efficient Learning for Multimodal Language Model
  • Compositional Chain-of-Thought Prompting for Large Multimodal Models
    :star:code
  • Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs
    :house:project
  • Honeybee: Locality-enhanced Projector for Multimodal LLM
    :star:codeLLM
  • HalluciDoctor: Mitigating Hallucinatory Toxicity in Visual Instruction Data
    :star:code
  • SEED-Bench: Benchmarking Multimodal Large Language Models
    :star:code
  • PerceptionGPT: Effectively Fusing Visual Perception into LLM
  • UniBind: LLM-Augmented Unified and Balanced Representation Space to Bind Them All
  • ModaVerse: Efficiently Transforming Modalities with LLMs
  • VCoder: Versatile Vision Encoders for Multimodal Large Language Models
    :star:code
    :house:project
  • mPLUG-Owl2: Revolutionizing Multi-modal Large Language Model with Modality Collaboration
  • MultiPLY: A Multisensory Object-Centric Embodied Large Language Model in 3D World
    :house:project大语言模型
  • RLHF-V: Towards Trustworthy MLLMs via Behavior Alignment from Fine-grained Correctional Human Feedback
    :star:code
  • DiffAgent: Fast and Accurate Text-to-Image API Selection with Large Language Model
    :star:code
    :thumbsup:摘要
  • Prompt Highlighter: Interactive Control for Multi-Modal LLMs
    :star:code
    :house:project
  • Auto MC-Reward: Automated Dense Reward Design with Large Language Models for Minecraft
    :house:project
  • General Object Foundation Model for Images and Videos at Scale
    :star:code
    :house:project
    :thumbsup:GLEE 华科与字节跳动联手打造全能目标感知基础模型
  • Link-Context Learning for Multimodal LLMs
    :star:codeLLMs
  • Cloud-Device Collaborative Learning for Multimodal Large Language Models
  • LocLLM: Exploiting Generalizable Human Keypoint Localization via Large Language Model
    :star:code
    :thumbsup:成果速览 | CVPR2024细粒度视觉感知多模态大模型Pink、LocLLM
  • Pink: Unveiling the Power of Referential Comprehension for Multi-modal LLMs
    :star:code
    :thumbsup:成果速览 | CVPR2024细粒度视觉感知多模态大模型Pink、LocLLM
  • LION: Empowering Multimodal Large Language Model with Dual-Level Visual Knowledge
    :star:code
    :house:projectMLLMs
  • GSVA: Generalized Segmentation via Multimodal Large Language Models
• VLN
  • Lookahead Exploration with Neural Radiance Representation for Continuous Vision-Language Navigation
    :star:code
    :thumbsup:VILP
  • Volumetric Environment Representation for Vision-Language Navigation
  • OVER-NAV: Elevating Iterative Vision-and-Language Navigation with Open-Vocabulary Detection and StructurEd Representation
  • Vision-and-Language Navigation via Causal Learning
    :star:code视觉和语言导航
• 视频语言
  • VidLA: Video-Language Alignment at Scale
  • SRTube: Video-Language Pre-Training with Action-Centric Video Tube Features and Semantic Role Labeling
  • VISTA-LLAMA: Reducing Hallucination in Video Language Models via Equal Distance to Visual Tokens
  • VideoLLM-online: Online Video Large Language Model for Streaming Video
    :house:project
• Visual Grounding
  • Naturally Supervised 3D Visual Grounding with Language-Regularized Concept Learners
  • MiKASA: Multi-Key-Anchor & Scene-Aware Transformer for 3D Visual Grounding
  • Viewpoint-Aware Visual Grounding in 3D Scenes
  • Improved Visual Grounding through Self-Consistent Explanations
  • Visual Programming for Zero-shot Open-Vocabulary 3D Visual Grounding
    :house:project
  • Towards CLIP-driven Language-free 3D Visual Grounding via 2D-3D Relational Enhancement and ConsistencyVisual Grounding
  • Marrying Hyperbolic Alignment with Explicit Semantic-Geometric Modeling for 3D Visual Grounding
  • Multi-Attribute Interactions Matter for 3D Visual Grounding
  • Investigating Compositional Challenges in Vision-Language Models for Visual Grounding
  • Separating the "Chirp" from the "Chat": Self-supervised Visual Grounding of Sound and Language
• 多模态模型
  • GLaMM: Pixel Grounding Large Multimodal Model
  • Monkey: Image Resolution and Text Label Are Important Things for Large Multi-modal Models
    :star:code
  • What If the TV Was Off? Examining Counterfactual Reasoning Abilities of Multi-modal Language Models
    :house:project
  • Multi-modal Learning for Geospatial Vegetation Forecasting
  • Multi-modal Instruction Tuned LLMs with Fine-grained Visual Perception
  • MP5: A Multi-modal Open-ended Embodied System in Minecraft via Active Perception
  • TRINS: Towards Multimodal Language Models that Can Read
  • Calibrating Multi-modal Representations: A Pursuit of Group Robustness without Annotations
• 视觉基础模型
  • Three Pillars improving Vision Foundation Model Distillation for Lidar
    :star:code
  • Bridging Remote Sensors with Multisensor Geospatial Foundation Models
  • Low-Resource Vision Challenges for Foundation Models
• 多视图理解
  • Learning to Select Views for Efficient Multi-View Understanding
    :star:code
• 视觉定位
  • Learning to Produce Semi-dense Correspondences for Visual Localization
    :star:code
  • PIN: Positional Insert Unlocks Object Localisation Abilities in VLMs定位
  • Efficient Privacy-Preserving Visual Localization Using 3D Ray Clouds

26.Information Security(信息安全)

• CGI-DM: Digital Copyright Authentication for Diffusion Models via Contrasting Gradient Inversion
  :star:code
• WateRF: Robust Watermarks in Radiance Fields for Protection of Copyrights
• 图像隐写术
  • Purified and Unified Steganographic Network
    :star:code
• 知识产权保护
  • Watermark-embedded Adversarial Examples for Copyright Protection against Diffusion Models
  • MAP: MAsk-Pruning for Source-Free Model Intellectual Property Protection
    :star:code
  • CPR: Retrieval Augmented Generation for Copyright Protection
  • VA3: Virtually Assured Amplification Attack on Probabilistic Copyright Protection for Text-to-Image Generative Models
    :star:code
  • Gaussian Shading: Provable Performance-Lossless Image Watermarking for Diffusion Models
• IP 保护
  • Steganographic Passport: An Owner and User Verifiable Credential for Deep Model IP Protection Without Retraining

25.Object Tracking(目标跟踪)

• 3D Feature Tracking via Event Camera
• Projecting Trackable Thermal Patterns for Dynamic Computer Vision
• ARTrackV2: Prompting Autoregressive Tracker Where to Look and How to Describe
• DPHMs: Diffusion Parametric Head Models for Depth-based Tracking
  :house:project
• NetTrack: Tracking Highly Dynamic Objects with a Net
  :star:code
• RTracker: Recoverable Tracking via PN Tree Structured Memory
• Context-Aware Integration of Language and Visual References for Natural Language Tracking
• CodedEvents: Optimal Point-Spread-Function Engineering for 3D-Tracking with Event Cameras
• SpatialTracker: Tracking Any 2D Pixels in 3D Space
  :star:code
• Learning Tracking Representations from Single Point Annotations
• 视觉目标跟踪
  • DiffusionTrack: Point Set Diffusion Model for Visual Object Tracking
  • OneTracker: Unifying Visual Object Tracking with Foundation Models and Efficient Tuning
  • HIPTrack: Visual Tracking with Historical Prompts
    :star:code
  • Single-Model and Any-Modality for Video Object Tracking
    :star:code
  • SDSTrack: Self-Distillation Symmetric Adapter Learning for Multi-Modal Visual Object Tracking
    :star:code
• 多目标跟踪
  • Multi-Object Tracking in the Dark
  • Towards Generalizable Multi-Object Tracking
  • ADA-Track: End-to-End Multi-Camera 3D Multi-Object Tracking with Alternating Detection and Association
  • DeconfuseTrack:Dealing with Confusion for Multi-Object Tracking
  • Delving into the Trajectory Long-tail Distribution for Muti-object Tracking
    :star:code
  • Self-Supervised Multi-Object Tracking with Path Consistency
  • DiffMOT: A Real-time Diffusion-based Multiple Object Tracker with Non-linear Prediction
    :star:code
    :house:project
  • iKUN: Speak to Trackers without Retraining
    :star:code
• 点跟踪
  • LEAP-VO: Long-term Effective Any Point Tracking for Visual Odometry

24.Machine Learning(机器学习)

• Molecular Data Programming: Towards Molecule Pseudo-labeling with Systematic Weak Supervision
  :thumbsup:摘要
• Improving Physics-Augmented Continuum Neural Radiance Field-Based Geometry-Agnostic System Identification with Lagrangian Particle Optimization
  :house:project
• Circuit Design and Efficient Simulation of Quantum Inner Product and Empirical Studies of Its Effect on Near-Term Hybrid Quantum-Classic Machine Learning
• 对抗
  • Infrared Adversarial Car Stickers
  • Robust Distillation via Untargeted and Targeted Intermediate Adversarial Samples
  • Revisiting Adversarial Training Under Long-Tailed Distributions
  • PAD: Patch-Agnostic Defense against Adversarial Patch Attacks
    :star:code
  • Structured Gradient-based Interpretations via Norm-Regularized Adversarial Training
  • MimicDiffusion: Purifying Adversarial Perturbation via Mimicking Clean Diffusion Model对抗性扰动
  • Towards Transferable Targeted 3D Adversarial Attack in the Physical World
  • Deep-TROJ: An Inference Stage Trojan Insertion Algorithm through Efficient Weight Replacement Attack攻击
  • Attack To Defend: Exploiting Adversarial Attacks for Detecting Poisoned Models
  • Re-thinking Data Availability Attacks Against Deep Neural Networks
  • SlowFormer: Adversarial Attack on Compute and Energy Consumption of Efficient Vision Transformers
  • Re-thinking Data Availablity Attacks Against Deep Neural Networks攻击
  • NAPGuard: Towards Detecting Naturalistic Adversarial Patches
  • Focus on Hiders: Exploring Hidden Threats for Enhancing Adversarial Training
  • Not All Prompts Are Secure: A Switchable Backdoor Attack Against Pre-trained Vision Transfomers后门攻击
  • Physical Backdoor: Towards Temperature-based Backdoor Attacks in the Physical World
  • Backdoor Defense via Test-Time Detecting and Repairing
  • Nearest Is Not Dearest: Towards Practical Defense against Quantization-conditioned Backdoor Attacks
  • Semantic-Aware Multi-Label Adversarial Attacks对抗攻击
  • Strong Transferable Adversarial Attacks via Ensembled Asymptotically Normal Distribution Learning
  • Improving Transferable Targeted Adversarial Attacks with Model Self-Enhancement对抗攻击
  • On the Robustness of Large Multimodal Models Against Image Adversarial Attacks
  • Incremental Residual Concept Bottleneck Models
  • Revisiting Adversarial Training at Scale
    :star:code
  • Language-Driven Anchors for Zero-Shot Adversarial Robustness零样本对抗
  • Transferable Structural Sparse Adversarial Attack Via Exact Group Sparsity Training
  • Learning to Transform Dynamically for Better Adversarial Transferability
  • Defense without Forgetting: Continual Adversarial Defense with Anisotropic & Isotropic Pseudo Replay
  • Boosting Adversarial Transferability by Block Shuffle and Rotation
    :star:code对抗性可转移性
  • MMCert: Provable Defense against Adversarial Attacks to Multi-modal Models
  • Pre-trained Model Guided Fine-Tuning for Zero-Shot Adversarial Robustness
    :thumbsup:VILP
  • Adversaral Doodles: Interpretable and Human-drawable Attacks Provide Describable Insights
  • PeerAiD: Improving Adversarial Distillation from a Specialized Peer Tutor
  • Revisiting Adversarial Training under Long-Tailed Distributions
    :star:code
  • Towards Fairness-Aware Adversarial Learning
  • Dispel Darkness for Better Fusion: A Controllable Visual Enhancer based on Cross-modal Conditional Adversarial Learning
  • Soften to Defend: Towards Adversarial Robustness via Self-Guided Label Refinement
  • Robust Overfitting Does Matter: Test-Time Adversarial Purification With FGSM
  • Boosting Adversarial Training via Fisher-Rao Norm-based Regularization
    :star:code
  • A Stealthy Wrongdoer: Feature-Oriented Reconstruction Attack against Split Learning攻击
  • 后门攻击
    • LOTUS: Evasive and Resilient Backdoor Attacks through Sub-Partitioning
      :star:code
    • Test-Time Backdoor Defense via Detecting and Repairing
    • Data Poisoning based Backdoor Attacks to Contrastive Learning
      :star:code
• 持续学习
  • RCL: Reliable Continual Learning for Unified Failure Detection
  • Consistent Prompting for Rehearsal-Free Continual Learning
  • Improving Plasticity in Online Continual Learning via Collaborative Learning
  • Boosting Continual Learning of Vision-Language Models via Mixture-of-Experts Adapters
    :star:code
  • Enhancing Visual Continual Learning with Language-Guided Supervision
  • Convolutional Prompting meets Language Models for Continual Learning
  • Resurrecting Old Classes with New Data for Exemplar-Free Continual Learning
  • Orchestrate Latent Expertise: Advancing Online Continual Learning with Multi-Level Supervision and Reverse Self-Distillation
  • InfLoRA: Interference-Free Low-Rank Adaptation for Continual Learning
  • Learning Equi-angular Representations for Online Continual Learning
  • BrainWash: A Poisoning Attack to Forget in Continual Learning
  • Adaptive VIO: Deep Visual-Inertial Odometry with Online Continual Learning持续学习
  • Traceable Federated Continual Learning
  • Interactive Continual Learning: Fast and Slow Thinking
• 增量学习
  • Towards Efficient Replay in Federated Incremental Learning
• 类增量学习
  • Dual-Consistency Model Inversion for Non-Exemplar Class Incremental Learning
  • Class Incremental Learning with Multi-Teacher Distillation
  • Dual-Enhanced Coreset Selection with Class-wise Collaboration for Online Blurry Class Incremental Learning
  • Generative Multi-modal Models are Good Class Incremental Learners
  • FCS: Feature Calibration and Separation for Non-Exemplar Class Incremental Learning
  • OrCo: Towards Better Generalization via Orthogonality and Contrast for Few-Shot Class-Incremental Learning
  • Long-Tail Class Incremental Learning via Independent Sub-prototype Construction
  • Gradient Reweighting: Towards Imbalanced Class-Incremental Learning
  • DYSON: Dynamic Feature Space Self-Organization for Online Task-Free Class Incremental Learning
  • NICE: Neurogenesis Inspired Contextual Encoding for Replay-free Class Incremental Learning
    :star:code
  • Expandable Subspace Ensemble for Pre-Trained Model-Based Class-Incremental Learning
    :star:code
  • Text-Enhanced Data-free Approach for Federated Class-Incremental Learning
    :star:code
  • Generative Multi-modal Models are Good Class-Incremental Learners
    :star:code
  • Task-Adaptive Saliency Guidance for Exemplar-free Class Incremental Learning
    :star:code
• 多任务
  • Masked AutoDecoder is Effective Multi-Task Vision Generalist
  • OmniVec2 - A Novel Transformer based Network for Large Scale Multimodal and Multitask Learning
  • Task-conditioned adaptation of visual features in multi-task policy learning
  • DiffusionMTL: Learning Multi-Task Denoising Diffusion Model from Partially Annotated Data
    :star:code
  • FedHCA2: Towards Hetero-Client Federated Multi-Task Learning
    :star:code
  • MTLoRA: A Low-Rank Adaptation Approach for Efficient Multi-Task Learning
  • Joint-Task Regularization for Partially Labeled Multi-Task Learning
  • Task-Conditioned Adaptation of Visual Features in Multi-Task Policy Learning
  • 多标签学习
    • View-Category Interactive Sharing Transformer for Incomplete Multi-View Multi-Label Learning
• 多视角学习
  • Rethinking Multi-view Representation Learning via Distilled Disentangling
    :star:code
• 元学习
  • FREE: Faster and Better Data-Free Meta-Learning
  • Improving Generalization via Meta-Learning on Hard Samples
• 联邦学习
  • An Aggregation-Free Federated Learning for Tackling Data Heterogeneity
  • Decentralized Directed Collaboration for Personalized Federated Learning
  • Rethinking the Representation in Federated Unsupervised Learning with Non-IID Data
  • Byzantine-robust Decentralized Federated Learning via Dual-domain Clustering and Trust Bootstrapping
  • FLHetBench: Benchmarking Device and State Heterogeneity in Federated Learning
  • Revamping Federated Learning Security from a Defender's Perspective: A Unified Defense with Homomorphic Encrypted Data Space
  • Unlocking the Potential of Prompt-Tuning in Bridging Generalized and Personalized Federated Learning
  • Mixed-Precision Quantization for Federated Learning on Resource-Constrained Heterogeneous Devices
  • FedSelect: Personalized Federated Learning with Customized Selection of Parameters for Fine-Tuning
  • Fair Federated Learning under Domain Skew with Local Consistency and Domain Diversity
    :star:code
  • Global and Local Prompts Cooperation via Optimal Transport for Federated Learning
  • PerAda: Parameter-Efficient Federated Learning Personalization with Generalization Guarantees
    :star:code
  • Relaxed Contrastive Learning for Federated Learning
  • DiPrompT: Disentangled Prompt Tuning for Multiple Latent Domain Generalization in Federated Learning
  • FedAS: Bridging Inconsistency in Personalized Federated Learning
    :star:code
  • Leak and Learn: An Attacker's Cookbook to Train Using Leaked Data from Federated Learning
  • Data Valuation and Detections in Federated Learning
    :star:code
  • An Upload-Efficient Scheme for Transferring Knowledge From a Server-Side Pre-trained Generator to Clients in Heterogeneous Federated Learning
    :star:code
  • Adaptive Hyper-graph Aggregation for Modality-Agnostic Federated Learning
  • FedUV: Uniformity and Variance for Heterogeneous Federated Learning
  • FedSOL: Stabilized Orthogonal Learning with Proximal Restrictions in Federated Learning
  • Communication-Efficient Federated Learning with Accelerated Client Gradient
• 强化学习
  • Improving Unsupervised Hierarchical Representation with Reinforcement Learning
  • AlignSAM: Aligning Segment Anything Model to Open Context via Reinforcement Learning强化学习
  • Training Diffusion Models Towards Diverse Image Generation with Reinforcement Learning
  • POCE: Primal Policy Optimization with Conservative Estimation for Multi-constraint Offline Reinforcement Learning
  • DMR: Decomposed Multi-Modality Representations for Frames and Events Fusion in Visual Reinforcement Learning
  • Learning to Control Camera Exposure via Reinforcement Learning
    :house:project
  • Regularized Parameter Uncertainty for Improving Generalization in Reinforcement Learning
  • Imitating Shortest Paths in Simulation Enables Effective Navigation and Manipulation in the Real World
    :house:project
• 多模态机器学习
  • DIEM: Decomposition-Integration Enhancing Multimodal Insights
• 迁移学习
  • Model Inversion Robustness: Can Transfer Learning Help?
  • Enhanced Motion-Text Alignment for Image-to-Video Transfer Learning
  • Structured Model Probing: Empowering Efficient Transfer Learning by Structured Regularization
  • UniPT: Universal Parallel Tuning for Transfer Learning with Efficient Parameter and Memory
    :star:code
  • Initialization Matters for Adversarial Transfer Learning
• 对比学习
  • Improving Graph Contrastive Learning via Adaptive Positive Sampling
  • MaskCLR: Attention-Guided Contrastive Learning for Robust Action Representation Learning
  • BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning
  • Universal Novelty Detection Through Adaptive Contrastive Learning
  • NoiseCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions in Diffusion Models
    :house:project
• 模仿学习
  • LASIL: Learner-Aware Supervised Imitation Learning For Long-term Microscopic Traffic Simulation
• 上下文学习
  • Skeleton-in-Context: Unified Skeleton Sequence Modeling with In-Context Learning
    :star:code
• 弱监督学习
  • Virtual Immunohistochemistry Staining for Histological Images Assisted by Weakly-supervised Learning
• 启示学习
  • One-Shot Open Affordance Learning with Foundation Models

23.Sound

• Hearing Anything Anywhere
  :house:project
• Unveiling the Power of Audio-Visual Early Fusion Transformers with Dense Interactions through Masked Modeling
• AV-RIR: Audio-Visual Room Impulse Response Estimation
  :tv:video
• DiffSal: Joint Audio and Video Learning for Diffusion Saliency Prediction
• Seeing and Hearing: Open-domain Visual-Audio Generation with Diffusion Latent Aligners
  :star:code
• Weakly-Supervised Audio-Visual Video Parsing with Prototype-based Pseudo-Labeling
• 视听对话
  • The Audio-Visual Conversational Graph: From an Egocentric-Exocentric Perspective
    :house:project
• 视听导航
  • RILA: Reflective and Imaginative Language Agent for Zero-Shot Semantic Audio-Visual Navigation
• 视听分割
  • Audio-Visual Segmentation via Unlabeled Frame Exploitation
  • Benchmarking Audio Visual Segmentation for Long-Untrimmed Videos
  • Cooperation Does Matter: Exploring Multi-Order Bilateral Relations for Audio-Visual Segmentation
  • Towards Robust Audiovisual Segmentation in Complex Environments with Quantization-based Semantic Decomposition
    :star:code
  • Unraveling Instance Associations: A Closer Look for Audio-Visual Segmentation
    :star:code
• 语音识别
  • A Study of Dropout-Induced Modality Bias on Robustness to Missing Video Frames for Audio-Visual Speech Recognition
    :star:code
• 语音定位
  • Learning to Visually Localize Sound Sources from Mixtures without Prior Source Knowledge
    :star:code
• 音-视语音表示学习
  • ES³: Evolving Self-Supervised Learning of Robust Audio-Visual Speech Representations
    :house:project
    :thumbsup:VILP
• 文本驱动的语音定位
  • T-VSL: Text-Guided Visual Sound Source Localization in Mixtures
• 从图像和语言提示合成音乐
  • MeLFusion: Synthesizing Music from Image and Language Cues using Diffusion Models
• 耳音频生成和定位
  • Cyclic Learning for Binaural Audio Generation and Localization
• 视频和音频同步
  • DiVAS: Video and Audio Synchronization with Dynamic Frame Rates
• 视听表征学习
  • Looking Similar Sounding Different: Leveraging Counterfactual Cross-Modal Pairs for Audiovisual Representation Learning
• 说话人检测
  • LoCoNet: Long-Short Context Network for Active Speaker Detection
    :star:code
• 音频描述
  • MM-Narrator: Narrating Long-form Videos with Multimodal In-Context Learning
    :house:project
• 视听语音翻译
  • AV2AV: Direct Audio-Visual Speech to Audio-Visual Speech Translation with Unified Audio-Visual Speech Representation
    :house:project视听语音到视听语音翻译

22.Deepfake Detection

• AVFF: Audio-Visual Feature Fusion for Video Deepfake Detection
• Preserving Fairness Generalization in Deepfake Detection
  :star:code
• Rethinking the Up-Sampling Operations in CNN-based Generative Network for Generalizable Deepfake Detection
  :star:code
• Exploiting Style Latent Flows for Generalizing Deepfake Video Detection
• LAA-Net: Localized Artifact Attention Network for Quality-Agnostic and Generalizable Deepfake Detection
• LAA-Net: Localized Artifact Attention Network for High-Quality Deepfakes Detection
• Transcending Forgery Specificity with Latent Space Augmentation for Generalizable Deepfake Detection
• Contrastive Learning for DeepFake Classification and Localization via Multi-Label Ranking
• 图像篡改检测
  • DiffForensics: Leveraging Diffusion Prior to Image Forgery Detection and Localization伪造图像检测
  • EditGuard: Versatile Image Watermarking for Tamper Localization and Copyright Protection
    :star:code用于篡改定位和版权保护的多功能图像水印
  • UnionFormer: Unified-Learning Transformer with Multi-View Representation for Image Manipulation Detection and Localization图像操作检测和定位
  • CORE-MPI: Consistency Object Removal with Embedding MultiPlane Image
• 合成图像检测
  • WinSyn: : A High Resolution Testbed for Synthetic Data
  • Forgery-aware Adaptive Transformer for Generalizable Synthetic Image Detection

21.Few/Zero-Shot Learning/DG/A(小/零样本/域泛化/域适应)

• Transductive Zero-Shot and Few-Shot CLIP
  :star:code
• DG
  • Disentangled Prompt Representation for Domain Generalization
  • A2XP: Towards Private Domain Generalization
    :star:code
    :house:project
  • PracticalDG: Perturbation Distillation on Vision-Language Models for Hybrid Domain Generalization
  • Towards Generalizing to Unseen Domains with Few Labels
  • Rethinking the Evaluation Protocol of Domain Generalization
  • Rethinking Multi-domain Generalization with A General Learning Objective
  • Unknown Prompt, the only Lacuna: Unveiling CLIP's Potential for Open Domain Generalization
    :star:code
  • Prompt-Driven Dynamic Object-Centric Learning for Single Domain Generalization
  • Efficiently Assemble Normalization Layers and Regularization for Federated Domain Generalization
• DA
  • Parameter Efficient Self-Supervised Geospatial Domain Adaptation
  • Learning CNN on ViT: A Hybrid Model to Explicitly Class-specific Boundaries for Domain Adaptation
  • Discriminative Pattern Calibration Mechanism for Source-Free Domain Adaptation
  • Understanding and Improving Source-free Domain Adaptation from a Theoretical Perspective
  • A Versatile Framework for Continual Test-Time Domain Adaptation: Balancing Discriminability and Generalizability
  • Domain-Agnostic Mutual Prompting for Unsupervised Domain Adaptation
  • Unveiling the Unknown: Unleashing the Power of Unknown to Known in Open-Set Source-Free Domain Adaptation
  • Unsupervised Video Domain Adaptation with Masked Pre-Training and Collaborative Self-Training
  • Revisiting the Domain Shift and Sample Uncertainty in Multi-source Active Domain Transfer
  • LEAD: Learning Decomposition for Source-free Universal Domain Adaptation
    :star:code
  • Split to Merge: Unifying Separated Modalities for Unsupervised Domain Adaptation
    :star:code
  • Source-Free Domain Adaptation with Frozen Multimodal Foundation Model
    :star:code
  • Universal Semi-Supervised Domain Adaptation by Mitigating Common-Class Bias
  • Unified Language-driven Zero-shot Domain Adaptation
    :house:project
• FSL
  • Adversarially Robust Few-shot Learning via Parameter Co-distillation of Similarity and Class Concept Learners
  • Descriptor and Word Soups: Overcoming the Parameter Efficiency Accuracy Tradeoff for Out-of-Distribution Few-shot Learning
  • Simple Semantic-Aided Few-Shot Learning
    :star:code
  • DeIL: Direct-and-Inverse CLIP for Open-World Few-Shot Learning
  • AMU-Tuning: Effective Logit Bias for CLIP-based Few-shot Learning
    :thumbsup:摘要
  • Discriminative Sample-Guided and Parameter-Efficient Feature Space Adaptation for Cross-Domain Few-Shot Learning
    :star:code
  • Flatten Long-Range Loss Landscapes for Cross-Domain Few-Shot Learning
  • Few-shot Learner Parameterization by Diffusion Time-steps
    :star:code
• ZSL
  • Progressive Semantic-Guided Vision Transformer for Zero-Shot Learning
  • Visual-Augmented Dynamic Semantic Prototype for Generative Zero-Shot Learning
    :thumbsup:提升生成式零样本学习能力，视觉增强动态语义原型方法
  • Context-based and Diversity-driven Specificity in Compositional Zero-Shot Learning
  • Troika: Multi-Path Cross-Modal Traction for Compositional Zero-Shot Learning
    :star:code
  • Improving Generalized Zero-Shot Learning by Exploring the Diverse Semantics from External Class Names

20.Optical Flow Estimation(光流估计)

• Efficient Meshflow and Optical Flow Estimation from Event Cameras
• UnSAMFlow: Unsupervised Optical Flow Guided by Segment Anything Model
  :star:code
• FlowTrack: Revisiting Optical Flow for Long-Range Dense Tracking
• FlowDiffuser: Advancing Optical Flow Estimation with Diffusion Models
• ADFactory: An Effective Framework for Generalizing Optical Flow with NeRF
• Dense Optical Tracking: Connecting the Dots
  :star:code
  :house:project光流
• MemFlow: Optical Flow Estimation and Prediction with Memory
  :star:code
• OCAI: Improving Optical Flow Estimation by Occlusion and Consistency Aware Interpolation
• 场景流
  • Bring Event into RGB and LiDAR: Hierarchical Visual-Motion Fusion for Scene Flow
  • ICP-Flow: LiDAR Scene Flow Estimation with ICP
• 3D 场景流估计
  • 3DSFLabelling: Boosting 3D Scene Flow Estimation by Pseudo Auto-labelling
  • DifFlow3D: Toward Robust Uncertainty-Aware Scene Flow Estimation with Iterative Diffusion-Based Refinement

19.Object Pose Estimation(物体姿态估计)

• 3D-LFM: Lifting Foundation Model
  :house:project
• Efficient Solution of Point-Line Absolute Pose
  :star:code
• Dual Pose-invariant Embeddings: Learning Category and Object-specific Discriminative Representations for Recognition and Retrieval
• DVMNet: Computing Relative Pose for Unseen Objects Beyond Hypotheses
  :star:code
• Dynamic Support Information Mining for Category-Agnostic Pose Estimation
• From Correspondences to Pose: Non-minimal Certifiably Optimal Relative Pose without Disambiguation
• 物体姿态估计
  • Object Pose Estimation via the Aggregation of Diffusion Features
    :star:code
  • NOPE: Novel Object Pose Estimation from a Single Image
    :star:code
  • GigaPose: Fast and Robust Novel Object Pose Estimation via One Correspondence
    :star:code
• 6DoF
  • HiPose: Hierarchical Binary Surface Encoding and Correspondence Pruning for RGB-D 6DoF Object Pose Estimation
  • Towards Co-Evaluation of Cameras HDR and Algorithms for Industrial-Grade 6DoF Pose Estimation
  • Confronting Ambiguity in 6D Object Pose Estimation via Score-Based Diffusion on SE(3)
  • SAM-6D: Segment Anything Model Meets Zero-Shot 6D Object Pose Estimation
    :star:code
  • FAR: Flexible Accurate and Robust 6DoF Relative Camera Pose Estimation
    :star:code
    :house:project
  • 6D-Diff: A Keypoint Diffusion Framework for 6D Object Pose Estimation
  • MatchU: Matching Unseen Objects for 6D Pose Estimation from RGB-D Images
  • Open-Vocabulary Object 6D Pose Estimation
  • FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objects
    :house:project
  • GenFlow: Generalizable Recurrent Flow for 6D Pose Refinement of Novel Objects
  • Open-vocabulary object 6D pose estimation
    :house:project
  • SecondPose: SE(3)-Consistent Dual-Stream Feature Fusion for Category-Level Pose Estimation
    :star:code
  • A Simple and Effective Point-based Network for Event Camera 6-DOFs Pose Relocalization
  • Instance-Adaptive and Geometric-Aware Keypoint Learning for Category-Level 6D Object Pose Estimation
  • MRC-Net: 6-DoF Pose Estimation with MultiScale Residual Correlation
  • Generalizing 6-DoF Grasp Detection via Domain Prior Knowledge
• 重识别
  • Magic Tokens: Select Diverse Tokens for Multi-modal Object Re-Identification
    :star:code
• 计数
  • Referring Expression Counting
  • Learning to Count without Annotations
  • Point, Segment and Count: A Generalized Framework for Object Counting
    :star:code
  • DAVE - A Detect-and-Verify Paradigm for Low-Shot Counting
    :star:code
  • Weakly Supervised Video Individual Counting

18.SLAM/AR/VR/Robotics(增强/虚拟现实/机器人)(机器人)

• Instance Tracking in 3D Scenes from Egocentric Videos
• VPR
  • TransLoc4D: Transformer-based 4D Radar Place Recognition
  • CricaVPR: Cross-image Correlation-aware Representation Learning for Visual Place Recognition
    :star:code
  • On the Estimation of Image-matching Uncertainty in Visual Place Recognition
  • Optimal Transport Aggregation for Visual Place Recognition
    :star:code
• 导航
  • Imagine Before Go: Self-Supervised Generative Map for Object Goal Navigation
    :thumbsup:VILP
  • Detours for Navigating Instructional Videos旅游视频导航
  • MemoNav: Working Memory Model for Visual Navigation
  • DiaLoc: An Iterative Approach to Embodied Dialog Localization
  • F$^3$Loc: Fusion and Filtering for Floorplan Localization
  • An Interactive Navigation Method with Effect-oriented Affordance交互式导航
• SLAM
  • SchurVINS: Schur Complement-Based Lightweight Visual Inertial Navigation System
    :star:code
  • SubT-MRS Dataset: Pushing SLAM Towards All-weather Environments
  • SNI-SLAM: Semantic Neural Implicit SLAM
  • Gaussian Splatting SLAM
    :star:code
    :house:project
  • SplaTAM: Splat, Track & Map 3D Gaussians for Dense RGB-D SLAM
    :house:project
  • NARUTO: Neural Active Reconstruction from Uncertain Target Observations
  • Benchmarking Implicit Neural Representation and Geometric Rendering in Real-Time RGB-D SLAM
  • Implicit Event-RGBD Neural SLAM
    :star:code
    :house:project
  • Multi-Session SLAM with Differentiable Wide-Baseline Pose Optimization
  • IBD-SLAM: Learning Image-Based Depth Fusion for Generalizable SLAM
  • Photo-SLAM: Real-time Simultaneous Localization and Photorealistic Mapping for Monocular Stereo and RGB-D Cameras
  • Loopy-SLAM: Dense Neural SLAM with Loop Closures
    :house:project
  • GS-SLAM: Dense Visual SLAM with 3D Gaussian Splatting
    :house:project
• 机器人
  • Retrieval-Augmented Embodied Agents
  • ManipLLM: Embodied Multimodal Large Language Model for Object-Centric Robotic Manipulation
  • SUGAR: Pre-training 3D Visual Representations for Robotics
    :house:project
  • Learning to navigate efficiently and precisely in real environments
  • Language-driven Grasp Detection
  • CyberDemo: Augmenting Simulated Human Demonstration for Real-World Dexterous Manipulation
    :house:project
  • Hierarchical Diffusion Policy for Kinematics-Aware Multi-Task Robotic Manipulation
    :star:code
  • Diffusion-EDFs:Bi-equivariant Denoising Generative Modeling on SE(3) for Visual Robotic Manipulation
    :star:code
    :house:project
  • Rapid Motor Adaptation for Robotic Manipulator Arms机器人机械臂
  • Generate Subgoal Images before Act: Unlocking the Chain-of-Thought Reasoning in Diffusion Model for Robot Manipulation with Multimodal Prompts
• Avatar(虚拟建模)
  • SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian Splatting
    :star:code
    :tv:video
  • MonoNPHM: Dynamic Head Reconstruction from Monocular Videos
  • Relightable and Animatable Neural Avatar from Sparse-View Video
    :house:project
  • Efficient 3D Implicit Head Avatar with Mesh-anchored Hash Table Blendshapes
    :star:code
  • Artist-Friendly Relightable and Animatable Neural Heads
  • GeneAvatar: Generic Expression-Aware Volumetric Head Avatar Editing from a Single Image
    :star:code
  • DiffusionAvatars: Deferred Diffusion for High-fidelity 3D Head Avatars
  • EMOPortraits: Emotion-enhanced Multimodal One-shot Head Avatars
  • Stratified Avatar Generation from Sparse Observations(https://zerg-overmind.github.io/)
    :tv:video
  • Real-Time Simulated Avatar from Head-Mounted Sensors
    :house:project
  • Animatable Gaussians: Learning Pose-dependent Gaussian Maps for High-fidelity Human Avatar Modeling
    :star:code
    :house:project
  • NECA: Neural Customizable Human Avatar
    :star:code
  • Make-Your-Anchor: A Diffusion-based 2D Avatar Generation Framework
    :star:code
  • GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians
    :star:code
    :house:project
  • Gaussian Head Avatar:Ultra High-fidelity Head Avatar via Dynamic Gaussians
    :star:code
    :house:project
  • GAvatar: Animatable 3D Gaussian Avatars with Implicit Mesh Learning
    :house:project
  • UltrAvatar: A Realistic Animatable 3D Avatar Diffusion Model with Authenticity Guided Textures
    :star:code
    :house:project
  • GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians
    :house:project
  • 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting
    :house:project3D动画
  • AttriHuman-3D: Editable 3D Human Avatar Generation with Attribute Decomposition and Indexing3D 人体头像生成
  • Human Gaussian Splatting: Real-time Rendering of Animatable Avatars
  • GoMAvatar: Efficient Animatable Human Modeling from Monocular Video Using Gaussians-on-Mesh
    :star:code
  • DiffAvatar: Simulation-Ready Garment Optimization with Differentiable Simulation
    :house:project
  • PEGASUS: Personalized Generative 3D Avatars with Composable Attributes
    :house:project
  • FlashAvatar: High-fidelity Head Avatar with Efficient Gaussian Embedding
    :house:project
  • MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model
    :house:project人体图像动画
  • Relightable Gaussian Codec Avatars
    :house:project
  • IntrinsicAvatar: Physically Based Inverse Rendering of Dynamic Humans from Monocular Videos via Explicit Ray Tracing
    :house:project
• 头发建模
  • MonoHair: High-Fidelity Hair Modeling from a Monocular Video
    :star:code
• 虚拟试穿
  • Texture-Preserving Diffusion Models for High-Fidelity Virtual Try-On
  • CAT-DM: Controllable Accelerated Virtual Try-on with Diffusion Model
  • M&M VTO: Multi-Garment Virtual Try-On and Editing
    :house:project
  • StableVITON: Learning Semantic Correspondence with Latent Diffusion Model for Virtual Try-On
    :star:code
  • PICTURE: PhotorealistIC virtual Try-on from UnconstRained dEsigns
    :house:project
• 抓取
  • MANUS: Markerless Grasp Capture using Articulated 3D Gaussians
• 卡通人物
  • Make-It-Vivid: Dressing Your Animatable Biped Cartoon Characters from Text
    :house:project

17.Automated Driving(自动驾驶)

• Bezier Everywhere All at Once: Learning Drivable Lanes as Bezier Graphs
  :star:code
• SparseOcc: Rethinking Sparse Latent Representation for Vision-Based Semantic Occupancy Prediction
• 自动驾驶
  • Accurate Training Data for Occupancy Map Prediction in Automated Driving Using Evidence Theory
  • VLP: Vision Language Planning for Autonomous Driving
  • Collaborative Semantic Occupancy Prediction with Hybrid Feature Fusion in Connected Automated Vehicles
  • DrivingGaussian: Composite Gaussian Splatting for Surrounding Dynamic Autonomous Driving Scenes
  • Diffusion-ES: Gradient-free Planning with Diffusion for Autonomous and Instruction-guided Driving
  • DualAD: Disentangling the Dynamic and Static World for End-to-End Driving
    :house:project
  • UniPAD: A Universal Pre-training Paradigm for Autonomous Driving
    :star:code
  • Generalized Predictive Model for Autonomous Driving
    :star:code
  • Cam4DOcc: Benchmark for Camera-Only 4D Occupancy Forecasting in Autonomous Driving Applications
  • ChatScene: Knowledge-Enabled Safety-Critical Scenario Generation for Autonomous Vehicles
  • Holistic Autonomous Driving Understanding by Bird's-Eye-View Injected Multi-Modal Large Models
  • LMDrive: Closed-Loop End-to-End Driving with Large Language Models
    :star:code
    :house:project
  • Feedback-Guided Autonomous Driving
  • PARA-Drive: Parallelized Architecture for Real-time Autonomous Driving
  • Is Ego Status All You Need for Open-Loop End-to-End Autonomous Driving
    :star:code
  • On the Road to Portability: Compressing End-to-End Motion Planner for Autonomous Driving
  • Visual Point Cloud Forecasting enables Scalable Autonomous Driving
    :star:code
  • Adaptive Fusion of Single-View and Multi-View Depth for Autonomous Driving
    :star:code
  • CLIP-BEVFormer: Enhancing Multi-View Image-Based BEV Detector with Ground Truth Flow
  • Physical 3D Adversarial Attacks against Monocular Depth Estimation in Autonomous Driving
  • AIDE: An Automatic Data Engine for Object Detection in Autonomous Driving
  • NeuRAD: Neural Rendering for Autonomous Driving
    :star:code
    :house:project
  • Driving into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving
    :star:code
    :house:project
  • Editable Scene Simulation for Autonomous Driving via Collaborative LLM-Agents
    :star:code
    :house:project
  • 3D LiDAR Mapping in Dynamic Environments using a 4D Implicit Neural Representation
    :star:code
  • PACER+: On-Demand Pedestrian Animation Controller in Driving Scenarios
    :star:code
    :tv:video
  • Bootstrapping Autonomous Driving Radars with Self-Supervised Learning
    :star:code
  • SynFog: A Photo-realistic Synthetic Fog Dataset based on End-to-end Imaging Simulation for Advancing Real-World Defogging in Autonomous Driving自动驾驶去雾
• 轨迹预测
  • Pose-Transformed Equivariant Network for 3D Point Trajectory Prediction
  • Adversarial Backdoor Attack by Naturalistic Data Poisoning on Trajectory Prediction in Autonomous Driving
  • CaDeT: a Causal Disentanglement Approach for Robust Trajectory Prediction in Autonomous Driving
  • Higher-order Relational Reasoning for Pedestrian Trajectory Prediction
  • Density-Adaptive Model Based on Motif Matrix for Multi-Agent Trajectory Prediction
  • GigaTraj: Predicting Long-term Trajectories of Hundreds of Pedestrians in Gigapixel Complex Scenes
  • ERMVP: Communication-Efficient and Collaboration-Robust Multi-Vehicle Perception in Challenging Environments
  • HPNet: Dynamic Trajectory Forecasting with Historical Prediction Attention
    :star:code
    :thumbsup:VILP
  • Adapting to Length Shift: FlexiLength Network for Trajectory Prediction
  • OOSTraj: Out-of-Sight Trajectory Prediction With Vision-Positioning Denoising
    :star:code
  • SocialCircle: Learning the Angle-based Social Interaction Representation for Pedestrian Trajectory Prediction行人轨迹预测
  • T4P: Test-Time Training of Trajectory Prediction via Masked Autoencoder and Actor-specific Token Memory
    :star:code
  • Self-Supervised Class-Agnostic Motion Prediction with Spatial and Temporal Consistency Regularizations
  • SmartRefine: An Scenario-Adaptive Refinement Framework for Efficient Motion Prediction
    :star:code
  • Producing and Leveraging Online Map Uncertainty in Trajectory Prediction
  • SingularTrajectory: Universal Trajectory Predictor Using Diffusion Model
    :star:code
  • Can Language Beat Numerical Regression? Language-Based Multimodal Trajectory Prediction
    :star:code
  • Quantifying Uncertainty in Motion Prediction with Variational Bayesian Mixture
    :star:code
• 车道线检测
  • LaneCPP: Continuous 3D Lane Detection using Physical Priors
  • Lane2Seq: Towards Unified Lane Detection via Sequence Generation
    :house:project
• 车载凝视估计
  • What Do You See in Vehicle? Comprehensive Vision Solution for In-Vehicle Gaze Estimation
    :house:project
• 3D Occupancy Prediction
  • COTR: Compact Occupancy TRansformer for Vision-based 3D Occupancy Prediction
    :star:code
  • SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction
    :star:code
  • StreamingFlow: Streaming Occupancy Forecasting with Asynchronous Multi-modal Data Streams via Neural Ordinary Differential Equation 车辆重识别
  • Day-Night Cross-domain Vehicle Re-identification

16.Point Cloud(点云)

• Single-View Scene Point Cloud Human Grasp Generation
• LTA-PCS: Learnable Task-Agnostic Point Cloud Sampling
• StraightPCF: Straight Point Cloud Filtering
• CurveCloudNet: Processing Point Clouds with 1D Structure
• Mitigating Object Dependencies: Improving Point Cloud Self-Supervised Learning through Object Exchange
• Learning SO(3)-Invariant Semantic Correspondence via Local Shape Transform
• Multiway Point Cloud Mosaicking with Diffusion and Global Optimization
  :tv:video
• Point Cloud Pre-training with Diffusion Models
• PBWR: Parametric-Building-Wireframe Reconstruction from Aerial LiDAR Point Clouds
• Unsupervised Occupancy Learning from Sparse Point Cloud
• TULIP: Transformer for Upsampling of LiDAR Point Clouds
• Draw Step by Step Like Human: Reconstructing CAD Construction Sequences from Point Clouds via Multimodal Diffusion点​​云重建 CAD
• Dynamic Adapter Meets Prompt Tuning: Parameter-Efficient Transfer Learning for Point Cloud Analysis
  :star:code
• Local-consistent Transformation Learning for Rotation-invariant Point Cloud Analysis
  :star:code
• Unsupervised Template-assisted Point Cloud Shape Correspondence Network
• GeoAuxNet: Towards Universal 3D Representation Learning for Multi-sensor Point Clouds
• Object Dynamics Modeling with Hierarchical Point Cloud-based Representations
• KPConvX: Modernizing Kernel Point Convolution with Kernel Attention
• 点云配准
  • Extend Your Own Correspondences: Unsupervised Distant Point Cloud Registration by Progressive Distance Extension
  • Inlier Confidence Calibration for Point Cloud Registration
  • ColorPCR: Color Point Cloud Registration with Multi-Stage Geometric-Color Fusion
  • Dynamic Cues-Assisted Transformer for Robust Point Cloud Registration
  • Learning Instance-Aware Correspondences for Robust Multi-Instance Point Cloud Registration in Cluttered Scenes
• 3D 点云
  • Hide in Thicket: Generating Imperceptible and Rational Adversarial Perturbations on 3D Point Clouds
    :star:code
  • Density-guided Translator Boosts Synthetic-to-Real Unsupervised Domain Adaptive Segmentation of 3D Point Clouds
    :star:code
    :thumbsup:摘要
  • Point2CAD: Reverse Engineering CAD Models from 3D Point Clouds
  • Text2Loc: 3D Point Cloud Localization from Natural Language
    :house:project
  • 3DInAction: Understanding Human Actions in 3D Point Clouds
    :star:code
  • Coupled Laplacian Eigenmaps for Locally-Aware 3D Rigid Point Cloud Matching
    :star:code
• 点云识别
  • X-3D: Explicit 3D Structure Modeling for Point Cloud Recognition
• 点云上采样
  • RepKPU: Point Cloud Upsampling with Kernel Point Representation and Deformation
  • A Conditional Denoising Diffusion Probabilistic Model for Point Cloud Upsampling
  • SPU-PMD: Self-Supervised Point Cloud Upsampling via Progressive Mesh Deformation
• 点云分割
  • Construct to Associate: Cooperative Context Learning for Domain Adaptive Point Cloud Segmentation
  • OneFormer3D: One Transformer for Unified Point Cloud Segmentation
  • 点云实例分割
    • FreePoint: Unsupervised Point Cloud Instance Segmentation
    • Spherical Mask: Coarse-to-Fine 3D Point Cloud Instance Segmentation with Spherical Representation3D 点云实例分割
  • 弱监督点云语义分割
    • Weakly Supervised Point Cloud Semantic Segmentation via Artificial Oracle
      :house:project
• 点云分析
  • TetraSphere: A Neural Descriptor for O(3)-Invariant Point Cloud Analysis
    :star:code
• 点云理解
  • Geometrically-driven Aggregation for Zero-shot 3D Point Cloud Understanding
• 点云生成
  • TIGER: Time-Varying Denoising Model for 3D Point Cloud Generation with Diffusion Process
    :star:code
• 点云去噪
  • Denoising Point Cloud in Latent Space via Graph Convolution and Invertible Neural Network
• 点云分类
  • CausalPC: Improving the Robustness of Point Cloud Classification by Causal Effect Identification
• 点云质量评估
  • Contrastive Pre-Training with Multi-View Fusion for No-Reference Point Cloud Quality Assessment无参考点云质量评估

15.Object Detection(目标检测)

• Semantic Line Combination Detector
• Language-conditioned Detection Transformer
• Unsupervised Salient Instance Detection
• Neural Exposure Fusion for High-Dynamic Range Object Detection
• LEOD: Label-Efficient Object Detection for Event Cameras
• SFOD: Spiking Fusion Object Detector
  :star:code
• Exploring Orthogonality in Open World Object Detection
  :star:code
• What How and When Should Object Detectors Update in Continually Changing Test Domains?
• Depth-Aware Concealed Crop Detection in Dense Agricultural Scenes
• SDDGR: Stable Diffusion-based Deep Generative Replay for Class Incremental Object Detection
• Salience DETR: Enhancing Detection Transformer with Hierarchical Salience Filtering Refinement
  :star:code
• Theoretically Achieving Continuous Representation of Oriented Bounding Boxes
  :star:code
• RadarDistill: Boosting Radar-based Object Detection Performance via Knowledge Distillation from LiDAR Features
• Boosting Object Detection with Zero-Shot Day-Night Domain Adaptation
  :star:code
• CRKD: Enhanced Camera-Radar Object Detection with Cross-modality Knowledge Distillation
  :house:project
• DETRs Beat YOLOs on Real-time Object Detection
  :house:project
• Hyperbolic Learning with Synthetic Captions for Open-World Detection
• Overload: Latency Attacks on Object Detection for Edge Devices
• YolOOD: Utilizing Object Detection Concepts for Multi-Label Out-of-Distribution Detection
• Active Domain Adaptation with False Negative Prediction for Object Detection
• RadSimReal: Bridging the Gap Between Synthetic and Real Data in Radar Object Detection With Simulation
  :star:code
• Active Object Detection with Knowledge Aggregation and Distillation from Large Models
• GLOW: Global Layout Aware Attacks on Object Detection
• Plug and Play Active Learning for Object Detection
  :star:code
• InstaGen: Enhancing Object Detection by Training on Synthetic Dataset
  :house:project
• Incorporating Geo-Diverse Knowledge into Prompting for Increased Geographical Robustness in Object Recognition
• Generating Enhanced Negatives for Training Language-Based Object Detectors
• SAR目标检测
  • Unleashing Channel Potential: Space-Frequency Selection Convolution for SAR Object Detection
• 3D目标检测
  • Pseudo Label Refinery for Unsupervised Domain Adaptation on Cross-dataset 3D Object Detection
  • PTT: Point-Trajectory Transformer for Efficient Temporal 3D Object Detection
  • Prompt3D: Random Prompt Assisted Weakly-Supervised 3D Object Detection
  • CaKDP: Category-aware Knowledge Distillation and Pruning Framework for Lightweight 3D Object Detection
  • Weakly Supervised Monocular 3D Detection with a Single-View Image
  • Weak-to-Strong 3D Object Detection with X-Ray Distillation
  • GAFusion: Adaptive Fusing LiDAR and Camera with Multiple Guidance for 3D Object Detection
  • BEVNeXt: Reviving Dense BEV Frameworks for 3D Object Detection
  • Towards Robust 3D Object Detection with LiDAR and 4D Radar Fusion in Various Weather Conditions
  • HUNTER: Unsupervised Human-centric 3D Detection via Transferring Knowledge from Synthetic Instances to Real Scenes
  • Commonsense Prototype for Outdoor Unsupervised 3D Object Detection
    :star:code
    :thumbsup:摘要
  • Multi-View Attentive Contextualization for Multi-View 3D Object Detection
  • BEVSpread: Spread Voxel Pooling for Bird's-Eye-View Representation in Vision-based Roadside 3D Object Detection
  • An Empirical Study of the Generalization Ability of Lidar 3D Object Detectors to Unseen Domains
  • SeaBird: Segmentation in Bird’s View with Dice Loss Improves Monocular 3D Detection of Large Objects
    :star:code
  • HINTED: Hard Instance Enhanced Detector with Mixed-Density Feature Fusion for Sparsely-Supervised 3D Object Detection
    :thumbsup:摘要
  • 3DiffTection: 3D Object Detection with Geometry-Aware Diffusion Features
    :house:project
  • UniMODE: Unified Monocular 3D Object Detection
  • Learning Occupancy for Monocular 3D Object Detection
    :star:code
  • CN-RMA: Combined Network with Ray Marching Aggregation for 3D Indoor Object Detection from Multi-view Images
    :star:code
  • VSRD: Instance-Aware Volumetric Silhouette Rendering for Weakly Supervised 3D Object Detection
    :star:code
  • Improving Distant 3D Object Detection Using 2D Box Supervision
  • SAFDNet: A Simple and Effective Network for Fully Sparse 3D Object Detection
    :star:code
  • Enhancing 3D Object Detection with 2D Detection-Guided Query Anchors
    :star:code
  • IS-Fusion: Instance-Scene Collaborative Fusion for Multimodal 3D Object Detection
    :star:code
  • RCBEVDet: Radar-camera Fusion in Bird's Eye View for 3D Object Detection
    :star:code
  • Decoupled Pseudo-labeling for Semi-Supervised Monocular 3D Object Detection
  • SeaBird: Segmentation in Bird's View with Dice Loss Improves Monocular 3D Detection of Large Objects
    :star:code
  • MonoCD: Monocular 3D Object Detection with Complementary Depths
    :star:code
• 小目标检测
  • Infrared Small Target Detection with Scale and Location Sensitivity
    :star:code
• 显著目标检测
  • VSCode: General Visual Salient and Camouflaged Object Detection with 2D Prompt Learning
    :star:code
• 定向目标检测
  • Rethinking Boundary Discontinuity Problem for Oriented Object Detection
    :star:code
  • PointOBB: Learning Oriented Object Detection via Single Point Supervision
    :star:code
  • 弱半监督定向目标检测
    • Relational Matching for Weakly Semi-Supervised Oriented Object Detection
• 小样本目标检测
  • SNIDA: Unlocking Few-Shot Object Detection with Non-linear Semantic Decoupling Augmentation
  • Few-Shot Object Detection with Foundation Models
• 域泛化目标检测
  • Unbiased Faster R-CNN for Single-source Domain Generalized Object Detection
  • Improving Single Domain-Generalized Object Detection: A Focus on Diversification and Alignment
• 域适应目标检测
  • D3T: Distinctive Dual-Domain Teacher Zigzagging Across RGB-Thermal Gap for Domain-Adaptive Object Detection
    :star:code
  • CAT: Exploiting Inter-Class Dynamics for Domain Adaptive Object Detection
• 开放式目标检测
  • Generative Region-Language Pretraining for Open-Ended Object Detection
• 半监督目标检测
  • A-Teacher: Asymmetric Network for 3D Semi-Supervised Object Detection
  • Sparse Semi-DETR: Sparse Learnable Queries for Semi-Supervised Object Detection
• 端到端目标检测
  • Point2RBox: Combine Knowledge from Synthetic Visual Patterns for End-to-end Oriented Object Detection with Single Point Supervision
    :star:code
• 开放词汇目标检测
  • Exploring Region-Word Alignment in Built-in Detector for Open-Vocabulary Object Detection
  • The devil is in the fine-grained details: Evaluating open-vocabulary object detectors for fine-grained understanding
    :house:project
  • Learning Background Prompts to Discover Implicit Knowledge for Open Vocabulary Object Detection
  • OVMR: Open-Vocabulary Recognition with Multi-Modal References开放词汇识别
  • SHiNe: Semantic Hierarchy Nexus for Open-vocabulary Object Detection
  • YOLO-World: Real-Time Open-Vocabulary Object Detection
    :star:code
  • Retrieval-Augmented Open-Vocabulary Object Detection
    :star:code
  • Taming Self-Training for Open-Vocabulary Object Detection
    :star:code
  • Scene-adaptive and Region-aware Multi-modal Prompt for Open Vocabulary Object Detection
  • DetCLIPv3: Towards Versatile Generative Open-vocabulary Object Detection
• 视频伪装目标检测
  • Endow SAM with Keen Eyes: Temporal-spatial Prompt Learning for Video Camouflaged Object Detection
• 基于事件的目标检测
  • EventDance: Unsupervised Source-free Cross-modal Adaptation for Event-based Object Recognition
  • Scene Adaptive Sparse Transformer for Event-based Object Detection
    :star:code
• 联合显著性目标检测
  • CosalPure: Learning Concept from Group Images for Robust Co-Saliency Detection
• 开集识别
  • From Coarse to Fine-Grained Open-Set Recognition
• 物体识别
  • Object Recognition as Next Token Prediction
    :star:code
• 目标发现
  • DIOD: Self-Distillation Meets Object Discovery
  • Adaptive Slot Attention: Object Discovery with Dynamic Slot Number
• 目标定位
  • CAM Back Again: Large Kernel CNNs from a Weakly Supervised Object Localization Perspective
    :star:code

14.Human Action Recognition(人体动作识别)

• STMixer: A One-Stage Sparse Action Detector
• Adapting Short-Term Transformers for Action Detection in Untrimmed Videos
• Skeleton2vec: A Self-supervised Learning Framework with Contextualized Target Representations for Skeleton Sequence
  :star:code
• Selective, Interpretable and Motion Consistent Privacy Attribute Obfuscation for Action Recognition
  :star:code
• Selective, Interpretable, and Motion Consistent Privacy Attribute Obfuscation for Action Recognition
  :star:code
  :house:project
• X-MIC: Cross-Modal Instance Conditioning for Egocentric Action Generalization
  :star:code
• LLMs are Good Action Recognizers
• Action Detection via an Image Diffusion Process
• Language Model Guided Interpretable Video Action Reasoning
  :star:code
• SoundingActions: Learning How Actions Sound from Narrated Egocentric Videos
  :house:project
• TIM: A Time Interval Machine for Audio-Visual Action Recognition
  :star:code
• VicTR: Video-conditioned Text Representations for Activity Recognition
• Action-slot: Visual Action-centric Representations for Multi-label Atomic Activity Recognition in Traffic Scenes
  :house:project
• Narrative Action Evaluation with Prompt-Guided Multimodal Interaction
  :star:code
• Align Before Adapt: Leveraging Entity-to-Region Alignments for Generalizable Video Action Recognition
• Modality-Collaborative Test-Time Adaptation for Action Recognition
• CPR-Coach: Recognizing Composite Error Actions based on Single-class Training
• 基于骨架的动作识别
  • BlockGCN: Redefine Topology Awareness for Skeleton-Based Action Recognition
• 基于事件的动作识别
  • ExACT: Language-guided Conceptual Reasoning and Uncertainty Estimation for Event-based Action Recognition and More
• 零样本动作识别
  • Part-aware Unified Representation of Language and Skeleton for Zero-shot Action Recognition
• 细粒度动作识别
  • PeVL: Pose-Enhanced Vision-Language Model for Fine-Grained Human Action Recognition
• 动作定位
  • Multiscale Vision Transformers Meet Bipartite Matching for Efficient Single-stage Action Localization
  • 时序动作定位
    • Realigning Confidence with Temporal Saliency Information for Point-Level Weakly-Supervised Temporal Action Localization
      :star:code
    • End-to-End Spatio-Temporal Action Localisation with Video Transformers
  • Test-Time Zero-Shot Temporal Action Localization
• 时序动作检测
  • Benchmarking the Robustness of Temporal Action Detection Models Against Temporal Corruptions
    :star:code
  • TE-TAD: Towards Full End-to-End Temporal Action Detection via Time-Aligned Coordinate Expression
  • End-to-End Temporal Action Detection with 1B Parameters Across 1000 Frames
    :star:code
  • Low-power, Continuous Remote Behavioral Localization with Event Cameras
    :house:project
  • Dual DETRs for Multi-Label Temporal Action Detection
• 动作质量评估
  • FineParser: A Fine-grained Spatio-temporal Action Parser for Human-centric Action Quality Assessment
• 群体活动识别
  • Bi-Causal: Group Activity Recognition via Bidirectional Causality
  • Learning from Synthetic Human Group Activities
• 人体动作理解
  • From Isolated Islands to Pangea: Unifying Semantic Space for Human Action Understanding
• 动作预期
  • Can’t make an Omelette without Breaking some Eggs: Plausible Action Anticipation using Large Video-Language Models
  • Uncertainty-aware Action Decoupling Transformer for Action Anticipation
• 行为定位
  • Low-power Continuous Remote Behavioral Localization with Event Cameras

13.Human Pose Estimation(人体姿态估计)

• CLOAF: CoLlisiOn-Aware Human Flow
• Meta-Point Learning and Refining for Category-Agnostic Pose Estimation
• SurMo: Surface-based 4D Motion Modeling for Dynamic Human Rendering
  :star:code
• GALA: Generating Animatable Layered Assets from a Single Scan
  :star:code
  :house:project
• ShapeMatcher: Self-Supervised Joint Shape Canonicalization Segmentation Retrieval and Deformation自监督关节形状规范化、分割、检索和变形
• 手部
  • Authentic Hand Avatar from a Phone Scan via Universal Hand Model
  • URHand: Universal Relightable Hands
    :house:project
  • OHTA: One-shot Hand Avatar via Data-driven Implicit Priors
    :star:code
    :house:project
  • BOTH2Hands: Inferring 3D Hands from Both Text Prompts and Body Dynamics
    :star:code
    :house:project
  • Reconstructing Hands in 3D with Transformers
  • 3D手部姿态估计
    • HOISDF: Constraining 3D Hand-Object Pose Estimation with Global Signed Distance Fields
      :star:code
    • Single-to-Dual-View Adaptation for Egocentric 3D Hand Pose Estimation
      :star:code
    • HandDiff: 3D Hand Pose Estimation with Diffusion on Image-Point Cloud
      :star:code
  • 手部网格重建
    • Complementing Event Streams and RGB Frames for Hand Mesh Reconstruction
      :star:code
      :house:project
    • HandBooster: Boosting 3D Hand-Mesh Reconstruction by Conditional Synthesis and Sampling of Hand-Object Interactions
      :star:code
  • 手部网格恢复
    • A Simple Baseline for Efficient Hand Mesh Reconstruction
      :star:code
      :house:project
    • HHMR: Holistic Hand Mesh Recovery by Enhancing the Multimodal Controllability of Graph Diffusion Models
      :house:project
  • 手部姿态跟踪
    • MS-MANO: Enabling Hand Pose Tracking with Biomechanical Constraints
  • 手部纹理重建
    • BiTT: Bi-directional Texture Reconstruction of Interacting Two Hands from a Single Image
      :star:code
      :house:project
  • 手势合成
    • Weakly-Supervised Emotion Transition Learning for Diverse 3D Co-speech Gesture Generation
      :house:project
    • ConvoFusion: Multi-Modal Conversational Diffusion for Co-Speech Gesture Synthesis
      :house:project
    • DiffSHEG: A Diffusion-Based Approach for Real-Time Speech-driven Holistic 3D Expression and Gesture Generation
      :house:project
    • EMAGE: Towards Unified Holistic Co-Speech Gesture Generation via Expressive Masked Audio Gesture Modeling
      :house:project
• 人体
  • LiveHPS: LiDAR-based Scene-level Human Pose and Shape Estimation in Free Environment
  • AiOS: All-in-One-Stage Expressive Human Pose and Shape Estimation
  • LPSNet: End-to-End Human Pose and Shape Estimation with Lensless Imaging
  • Dynamic Inertial Poser (DynaIP): Part-Based Motion Dynamics Learning for Enhanced Human Pose Estimation with Sparse Inertial Sensors
  • Fast Adaptation for Human Pose Estimation via Meta-Optimization
  • RAM-Avatar: Real-time Photo-Realistic Avatar from Monocular Videos with Full-body Control
    :star:code
  • SDPose: Tokenized Pose Estimation via Circulation-Guide Self-Distillation
    :star:code
  • 多人姿势估计
    • DiffusionRegPose: Enhancing Multi-Person Pose Estimation using a Diffusion-Based End-to-End Regression Approach
    • RTMO: Towards High-Performance One-Stage Real-Time Multi-Person Pose Estimation
  • 3D 人体
    • Person-in-WiFi 3D: End-to-End Multi-Person 3D Pose Estimation with Wi-Fi
    • Cross-view and Cross-pose Completion for 3D Human Understanding
    • TexVocab: Texture Vocabulary-conditioned Human Avatars
      :house:project
    • MonoDiff: Monocular 3D Object Detection and Pose Estimation with Diffusion Models
    • ChatPose: Chatting about 3D Human Pose
      :house:project
    • SelfPose3d: Self-Supervised Multi-Person Multi-View 3d Pose Estimation
      :star:code
    • Attention-Propagation Network for Egocentric Heatmap to 3D Pose Lifting
    • FutureHuman3D: Forecasting Complex Long-Term 3D Human Behavior from Video Observations
      :house:project
    • FinePOSE: Fine-Grained Prompt-Driven 3D Human Pose Estimation via Diffusion Models
    • PoseIRM: Enhance 3D Human Pose Estimation on Unseen Camera Settings via Invariant Risk Minimization
    • Score-Guided Diffusion for 3D Human Recovery
      :star:code
    • A Dual-Augmentor Framework for Domain Generalization in 3D Human Pose Estimation
    • KTPFormer: Kinematics and Trajectory Prior Knowledge-Enhanced Transformer for 3D Human Pose Estimation
      :star:code
    • Multiple View Geometry Transformers for 3D Human Pose Estimation
      :star:code
    • Normalizing Flows on the Product Space of SO(3) Manifolds for Probabilistic Human Pose Modeling
    • Multi-agent Long-term 3D Human Pose Forecasting via Interaction-aware Trajectory Conditioning
      :star:code
    • EventEgo3D: 3D Human Motion Capture from Egocentric Event Streams
      :house:project
    • 3D Human Pose Perception from Egocentric Stereo Videos
      :star:code
      :house:project
    • Forecasting of 3D Whole-body Human Poses with Grasping Objects3D 全身人体姿势
    • BodyMAP -- Jointly Predicting Body Mesh and 3D Applied Pressure Map for People in Bed
      :star:code
      :house:project
    • MeshPose: Unifying DensePose and 3D Body Mesh Reconstruction
    • Exploring Vision Transformers for 3D Human Motion-Language Models with Motion Patches
    • Hourglass Tokenizer for Efficient Transformer-Based 3D Human Pose Estimation
      :star:code
      :thumbsup:让视频姿态Transformer变得飞速，北大提出高效三维人体姿态估计框架HoT
    • Optimizing Diffusion Noise Can Serve As Universal Motion Priors
      :star:code
      :house:project
    • En3D: An Enhanced Generative Model for Sculpting 3D Humans from 2D Synthetic Data
      :star:code
      :house:project
• 人体网格恢复/重建
  • DPMesh: Exploiting Diffusion Prior for Occluded Human Mesh Recovery
  • Instance-aware Contrastive Learning for Occluded Human Mesh Reconstruction
  • PostureHMR: Posture Transformation for 3D Human Mesh Recovery
  • Probabilistic Human Mesh Estimation with Hypothesis Scoring
  • KITRO:Refining Human Mesh by 2D Clues and Kinematic-tree Rotation
    :star:code
  • Semantic Human Mesh Reconstruction with Textures
    :house:project
  • TokenHMR: Advancing Human Mesh Recovery with a Tokenized Pose Representation
    :house:project
  • SCULPT: Shape-Conditioned Unpaired Learning of Pose-dependent Clothed and Textured Human Meshes人体网格
  • R-Cyclic Diffuser: Reductive and Cyclic Latent Diffusion for 3D Clothed Human Digitalization
  • DiffusionPoser: Real-time Human Motion Reconstruction From Arbitrary Sparse Sensors Using Autoregressive Diffusion
    :house:project
  • Synergistic Global-space Camera and Human Reconstruction from Videos
  • SiTH: Single-view Textured Human Reconstruction with Image-Conditioned Diffusion
  • SIFU: Side-view Conditioned Implicit Function for Real-world Usable Clothed Human Reconstruction
• 动作捕捉
  • ProxyCap: Real-time Monocular Full-body Capture in World Space via Human-Centric Proxy-to-Motion Learning
    :house:project
  • Egocentric Whole-Body Motion Capture with FisheyeViT and Diffusion-Based Motion Refinement
  • Capturing Closely Interacted Two-Person Motions with Reaction Priors
    :house:project
  • Loose Inertial Poser: Motion Capture with IMU-attached Loose-Wear Jacket
    :star:code
  • Mocap Everyone Everywhere: Lightweight Motion Capture With Smartwatches and a Head-Mounted Camera
    :star:code
    :house:project
• 3D人体生成
  • HumanNorm: Learning Normal Diffusion Model for High-quality and Realistic 3D Human Generation
    :star:code
    :house:project
  • FlowMDM: Seamless Human Motion Composition with Blended Positional Encodings
    :star:code
    :house:project人体运动合成
  • HumanRef: Single Image to 3D Human Generation via Reference-Guided Diffusion
    :house:project
  • Gaussian Shell Maps for Efficient 3D Human Generation
• 语音驱动的人体动画
  • Emotional Speech-driven 3D Body Animation via Disentangled Latent Diffusion
    :star:code
    :house:project
• 文本提示的人体动画
  • HOIAnimator: Generating Text-prompt Human-object Animations using Novel Perceptive Diffusion Models
• 手语翻译
  • LLMs are Good Sign Language Translators
  • Neural Sign Actors: A Diffusion Model for 3D Sign Language Production from Text
• 3D姿势迁移
  • Towards Robust 3D Pose Transfer with Adversarial Learning
  • Cinematic Behavior Transfer via NeRF-based Differentiable Filming
• 人体重建
  • MultiPly: Reconstruction of Multiple People from Monocular Video in the Wild
  • Closely Interactive Human Reconstruction with Proxemics and Physics-Guided Adaption
    :star:code
  • Joint Reconstruction of 3D Human and Object via Contact-Based Refinement Transformer
    :star:code
  • HiLo: Detailed and Robust 3D Clothed Human Reconstruction with High-and Low-Frequency Information of Parametric Models
  • ANIM: Accurate Neural Implicit Model for Human Reconstruction from a single RGB-D Image
  • Diffusion-FOF: Single-View Clothed Human Reconstruction via Diffusion-Based Fourier Occupancy Field
  • VS: Reconstructing Clothed 3D Human from Single Image via Vertex Shift
    :star:code
  • WHAM: Reconstructing World-grounded Humans with Accurate 3D Motion
    :house:project3D 运动重建
• 类别无关的姿势估计
  • ESCAPE: Encoding Super-keypoints for Category-Agnostic Pose Estimation
• 视频估计人体动力学
  • PhysPT: Physics-aware Pretrained Transformer for Estimating Human Dynamics from Monocular Videos
• 人体姿势回归
  • Video-Based Human Pose Regression via Decoupled Space-Time Aggregation
    :star:code
• 3D人体模型
  • GauHuman: Articulated Gaussian Splatting from Monocular Human Videos
    :star:code
    :house:project
• 人体生成
  • FairRAG: Fair Human Generation via Fair Retrieval Augmentation
  • HumanGaussian: Text-Driven 3D Human Generation with Gaussian Splatting
    :house:project文本驱动 3D 人体生成
  • Joint2Human: High-Quality 3D Human Generation via Compact Spherical Embedding of 3D Joints
    :house:project
  • MoMask: Generative Masked Modeling of 3D Human Motions
    :house:project3D 人体运动
• 人体运动理解
  • HumMUSS: Human Motion Understanding using State Space Models
• 人体形状
  • Distilling CLIP with Dual Guidance for Learning Discriminative Human Body Shape Representation
• 舞蹈生成
  • DisCo: Disentangled Control for Realistic Human Dance Generation
    :house:project
  • POPDG: Popular 3D Dance Generation with PopDanceSet
  • DanceCamera3D: 3D Camera Movement Synthesis with Music and Dance
    :star:code
  • Lodge: A Coarse to Fine Diffusion Network for Long Dance Generation Guided by the Characteristic Dance Primitives
    :house:project
  • Bidirectional Autoregessive Diffusion Model for Dance Generation

12.Video

• Learning from One Continuous Video Stream
• Deep Video Inverse Tone Mapping Based on Temporal Clues
• VTimeLLM: Empower LLM to Grasp Video Moments
• Combining Frame and GOP Embeddings for Neural Video Representation
• Learning to Predict Activity Progress by Self-Supervised Video Alignment
• CoDeF: Content Deformation Fields for Temporally Consistent Video Processing
• vid-TLDR: Training Free Token Merging for Light-weight Video Transformer
  :star:code
• Video2Game: Real-time Interactive Realistic and Browser-Compatible Environment from a Single Video
  :star:code
• Dancing with Still Images: Video Distillation via Static-Dynamic Disentanglement
• Understanding Video Transformers via Universal Concept Discovery
• Video Recognition in Portrait Mode
  :house:project
• VideoRF: Rendering Dynamic Radiance Fields as 2D Feature Video Streams
  :house:project
• Just Add π! Pose Induced Video Transformers for Understanding Activities of Daily Living
  :star:code
• A Simple Recipe for Contrastively Pre-training Video-First Encoders Beyond 16 Frames
• [Reliable Video Teller via Equal Distance to Visual Tokens]
• Vista-LLaMA: Reliable Video Narrator via Equal Distance to Visual Tokens
  :house:project
• Towards HDR and HFR Video from Rolling-Mixed-Bit Spikings
• Physics-guided Shape-from-Template: Monocular Video Perception through Neural Surrogate Models
• 睡眠监测
  • SleepVST: Sleep Staging from Near-Infrared Video Signals using Pre-Trained Transformers
• 视频理解
  • Compositional Video Understanding with Spatiotemporal Structure-based Transformers
  • Action Scene Graphs for Long-Form Understanding of Egocentric Videos
  • HIG: Hierarchical Interlacement Graph Approach to Scene Graph Generation in Video Understanding
  • A Backpack Full of Skills: Egocentric Video Understanding with Diverse Task Perspectives
    :house:project
  • Koala: Key Frame-Conditioned Long Video-LLM
  • MA-LMM: Memory-Augmented Large Multimodal Model for Long-Term Video Understanding
    :star:code
  • Abductive Ego-View Accident Video Understanding for Safe Driving Perception
    :house:project
  • OmniVid: A Generative Framework for Universal Video Understanding
    :star:code
  • A Unified Framework for Human-centric Point Cloud Video Understanding
  • Bridging the Gap: A Unified Video Comprehension Framework for Moment Retrieval and Highlight Detection
  • MovieChat: From Dense Token to Sparse Memory for Long Video Understanding
    :house:project
  • TimeChat: A Time-sensitive Multimodal Large Language Model for Long Video Understanding
    :star:code
  • Chat-UniVi: Unified Visual Representation Empowers Large Language Models with Image and Video Understanding
    :star:code
• 视频摘要
  • Previously on ... From Recaps to Story Summarization
    :house:project
  • Scaling Up Video Summarization Pretraining with Large Language Models
  • CSTA: CNN-based Spatiotemporal Attention for Video Summarization
    :star:code
• 视频重建
  • HDRFlow: Real-Time HDR Video Reconstruction with Large Motions
    :star:code
• 视频表示
  • DS-NeRV: Implicit Neural Video Representation with Decomposed Static and Dynamic Codes
    :house:project
• 视频判读
  • Visual Objectification in Films: Towards a New AI Task for Video Interpretation
• 电影描述
  • MICap: A Unified Model for Identity-Aware Movie Descriptions
    :house:project
• 视频监控
  • Towards Surveillance Video-and-Language Understanding: New Dataset Baselines and Challenges
    :sunflower:dataset
• 视频预测
  • Video Prediction by Modeling Videos as Continuous Multi-Dimensional Processes
  • ExtDM: Distribution Extrapolation Diffusion Model for Video Prediction
    :star:code
    :house:project
• 视频稳定
  • Harnessing Meta-Learning for Improving Full-Frame Video Stabilization
  • 3D Multi-frame Fusion for Video Stabilization
• 视频识别
  • OST: Refining Text Knowledge with Optimal Spatio-Temporal Descriptor for General Video Recognition
    :star:code
    :house:project
• 视频对话
  • BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning
    :star:code
• 视频重照明
  • Real-time 3D-aware Portrait Video Relighting
• 视频和谐化
  • Video Harmonization with Triplet Spatio-Temporal Variation Patterns
    :thumbsup:VILP
• 视频帧插值
  • Video Frame Interpolation via Direct Synthesis with the Event-based Reference
  • IQ-VFI: Implicit Quadratic Motion Estimation for Video Frame Interpolation
  • EVS-assisted Joint Deblurring Rolling-Shutter Correction and Video Frame Interpolation through Sensor Inverse Modeling
  • TTA-EVF: Test-Time Adaptation for Event-based Video Frame Interpolation via Reliable Pixel and Sample Estimation
  • Sparse Global Matching for Video Frame Interpolation with Large Motion
    :star:code
  • Perception-Oriented Video Frame Interpolation via Asymmetric Blending
    :star:code
    :thumbsup:视频插帧视觉效果新突破！上海交大提出PerVFI，视频插帧新范式
  • SportsSloMo: A New Benchmark and Baselines for Human-centric Video Frame Interpolation
    :house:project
• 视频主题交换
  • VideoSwap: Customized Video Subject Swapping with Interactive Semantic Point Correspondence
    :house:project
• 视频异常检测
  • Open-Vocabulary Video Anomaly Detection
  • Multi-Scale Video Anomaly Detection by Multi-Grained Spatio-Temporal Representation Learning
  • Harnessing Large Language Models for Training-free Video Anomaly Detection
    :star:code
  • Collaborative Learning of Anomalies with Privacy (CLAP) for Unsupervised Video Anomaly Detection: A New Baseline
    :star:code
  • Prompt-Enhanced Multiple Instance Learning for Weakly Supervised Video Anomaly Detection
  • MULDE: Multiscale Log-Density Estimation via Denoising Score Matching for Video Anomaly Detection
  • PREGO: Online Mistake Detection in PRocedural EGOcentric Videos
  • Self-Distilled Masked Auto-Encoders are Efficient Video Anomaly Detectors
    :star:code
  • Text Prompt with Normality Guidance for Weakly Supervised Video Anomaly Detection
  • GlitchBench: Can Large Multimodal Models Detect Video Game Glitches?
    :house:project大型多模态模型能否检测视频游戏故障
• 视频场景检测
  • Neighbor Relations Matter in Video Scene Detection
• 视频镜像检测
  • Effective Video Mirror Detection with Inconsistent Motion Cues
• 自动生成电影预告片
  • Towards Automated Movie Trailer Generation
• 视频对话式音乐推荐系统
  • MuseChat: A Conversational Music Recommendation System for Videos
• Video Paragraph Grounding
  • Siamese Learning with Joint Alignment and Regression for Weakly-Supervised Video Paragraph Grounding
• video Grounding
  • SnAG: Scalable and Accurate Video Grounding
    :star:code
  • Context-Guided Spatio-Temporal Video Grounding
    :star:code
  • Video-GroundingDINO: Towards Open-Vocabulary Spatio-Temporal Video Grounding
  • What When and Where? Self-Supervised Spatio-Temporal Grounding in Untrimmed Multi-Action Videos from Narrated Instructions

11.3D

• Rapid 3D Model Generation with Intuitive 3D Input
• Instantaneous Perception of Moving Objects in 3D
• NEAT: Distilling 3D Wireframes from Neural Attraction Fields
  :star:code
• Sculpting Holistic 3D Representation in Contrastive Language-Image-3D Pre-training
• LowRankOcc: Tensor Decomposition and Low-Rank Recovery for Vision-based 3D Semantic Occupancy Prediction
• TexOct: Generating Textures of 3D Models with Octree-based Diffusion
• Unsupervised 3D Structure Inference from Category-Specific Image Collections
• Garment Recovery with Shape and Deformation Priors
• ULIP-2: Towards Scalable Multimodal Pre-training for 3D Understanding
  :star:code
• CAGE: Controllable Articulation GEneration
  :star:code
  :house:project3D
• Sparse views, Near light: A practical paradigm for uncalibrated point-light photometric stereo
• Dispersed Structured Light for Hyperspectral 3D Imaging
• G-FARS: Gradient-Field-based Auto-Regressive Sampling for 3D Part Grouping
  :star:code
• Wonder3D: Single Image to 3D using Cross-Domain Diffusion
  :house:project
• UniGarmentManip: A Unified Framework for Category-Level Garment Manipulation via Dense Visual Correspondence
  :star:code服装操作
• GoMVS: Geometrically Consistent Cost Aggregation for Multi-View Stereo
  :star:code
  :star:code
• EfficientDreamer: High-Fidelity and Robust 3D Creation via Orthogonal-view Diffusion Priors
  :house:project
• MVD-Fusion: Single-view 3D via Depth-consistent Multi-view Generation
  :house:project
• Digital Life Project: Autonomous 3D Characters with Social Intelligence
  :house:project
• Image Sculpting: Precise Object Editing with 3D Geometry Control
  :house:project
• TutteNet: Injective 3D Deformations by Composition of 2D Mesh Deformations
• Building a Strong Pre-Training Baseline for Universal 3D Large-Scale Perception
• GaussianEditor: Swift and Controllable 3D Editing with Gaussian Splatting
  :star:code
  :house:project
• SHAP-EDITOR: Instruction-Guided Latent 3D Editing in Seconds
• ESR-NeRF: Emissive Source Reconstruction Using LDR Multi-view Images
• Differentiable Display Photometric Stereo
• ConsistNet: Enforcing 3D Consistency for Multi-view Images Diffusion
  :star:code
  :house:project
• Improving Semantic Correspondence with Viewpoint-Guided Spherical Maps
• REACTO: Reconstructing Articulated Objects from a Single Video
  :star:code
• Low-Latency Neural Stereo Streaming
• Unsigned Orthogonal Distance Fields: An Accurate Neural Implicit Representation for Diverse 3D Shapes
• Spectrum AUC Difference (SAUCD): Human-aligned 3D Shape Evaluation
  :house:project
• Wired Perspectives: Multi-View Wire Art Embraces Generative AI
  :star:code
  :house:project
• Memory-based Adapters for Online 3D Scene Perception
  :star:code
• FastMAC: Stochastic Spectral Sampling of Correspondence Graph
  :star:code
• One-2-3-45++: Fast Single Image to 3D Objects with Consistent Multi-View Generation and 3D Diffusion
  :star:code
  :house:project
• PonderV2: Pave the Way for 3D Foundation Model with A Universal Pre-training Paradigm
  :house:project
• CityDreamer: Compositional Generative Model of Unbounded 3D Cities
  :house:project
• EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AI
  :star:code
• Mosaic-SDF for 3D Generative Models
  :house:project
• Federated Online Adaptation for Deep Stereo
  :house:project
• ControlRoom3D: Room Generation using Semantic Proxy Rooms
• 三维视觉
  • Situational Awareness Matters in 3D Vision Language Reasoning
    :house:project
  • DUSt3R: Geometric 3D Vision Made Easy
  • Towards 3D Vision with Low-Cost Single-Photon Cameras
• 三维重建
  • 3D Neural Edge Reconstruction
  • 3DFIRES: Few Image 3D REconstruction for Scenes with Hidden Surfaces
    :house:project
    :tv:video
  • PanoRecon: Real-Time Panoptic 3D Reconstruction from Monocular Video
  • NeRSP: Neural 3D Reconstruction for Reflective Objects with Sparse Polarized Images
  • NTO3D: Neural Target Object 3D Reconstruction with Segment Anything
  • pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction
  • ReconFusion: 3D Reconstruction with Diffusion Priors
  • VGGSfM: Visual Geometry Grounded Deep Structure From Motion
    :star:code
    :house:project
  • Slice3D: Multi-Slice Occlusion-Revealing Single View 3D Reconstruction
    :house:project
  • GenNBV: Generalizable Next-Best-View Policy for Active 3D Reconstruction
  • Coherence As Texture - Passive Textureless 3D Reconstruction by Self-interference
    :star:code
  • Structure-Aware Sparse-View X-ray 3D Reconstruction
    :star:code
    :thumbsup:如何给 NeRF 开透视眼？
  • Triplane Meets Gaussian Splatting: Fast and Generalizable Single-View 3D Reconstruction with Transformers
    :house:project
  • Carve3D: Improving Multi-view Reconstruction Consistency for Diffusion Models with RL Finetuning
    :star:code
    :house:project多视图重建
  • PlatoNeRF: 3D Reconstruction in Plato’s Cave via Single-View Two-Bounce Lidar
    :house:project
  • WonderJourney: Going from Anywhere to Everywhere
    :house:project
  • Living Scenes: Multi-object Relocalization and Reconstruction in Changing 3D Environments
    :star:code
    :house:project
  • DiffHuman: Probabilistic Photorealistic 3D Reconstruction of Humans
  • IPoD: Implicit Field Learning with Point Diffusion for Generalizable 3D Object Reconstruction from Single RGB-D Images
    :star:code
  • Splatter Image: Ultra-Fast Single-View 3D Reconstruction
    :star:code
    :house:project
  • PlatoNeRF: 3D Reconstruction in Plato's Cave via Single-View Two-Bounce Lidar
    :star:code
    :house:project
  • MicroDiffusion: Implicit Representation-Guided Diffusion for 3D Reconstruction from Limited 2D Microscopy Projections
    :star:code
  • ZeroShape: Regression-based Zero-shot Shape Reconstruction
    :star:code
    :house:project
  • DITTO: Dual and Integrated Latent Topologies for Implicit 3D Reconstruction
  • G3DR: Generative 3D Reconstruction in ImageNet
    :star:code
    :house:project
  • 3DFIRES: Few Image 3D REconstruction for Scenes with Hidden Surface
    :star:code
  • Bayesian Diffusion Models for 3D Shape Reconstruction
  • RNb-NeuS: Reflectance and Normal-based Multi-View 3D Reconstruction
  • ZeroRF: Fast Sparse View 360deg Reconstruction with Zero Pretraining
    :house:project视图 360° 重建
• 表面重建
  • SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integration
  • MVCPS-NeuS: Multi-view Constrained Photometric Stereo for Neural Surface Reconstruction
  • MorpheuS: Neural Dynamic 360deg Surface Reconstruction from Monocular RGB-D Video
    :star:code
    :house:project
  • UFORecon: Generalizable Sparse-View Surface Reconstruction from Arbitrary and UnFavOrable Data Sets
    :star:code
    :star:code
  • UFORecon: Generalizable Sparse-View Surface Reconstruction from Arbitrary and Unfavorable Sets
    :star:code
• 三维网格重建
  • SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering
    :house:project
• 三维形状
  • GPLD3D: Latent Diffusion of 3D Shape Generative Models by Enforcing Geometric and Physical Priors
  • TAMM: TriAdapter Multi-Modal Learning for 3D Shape Understanding
    :star:code
  • Doodle Your 3D: From Abstract Freehand Sketches to Precise 3D Shapes
    :house:project
  • ShapeWalk: Compositional Shape Editing Through Language-Guided Chains
    :star:code
    :house:project
  • Spectral Meets Spatial: Harmonising 3D Shape Matching and Interpolation
  • Open-Vocabulary 3D Scene Graphs from Point Clouds with Queryable Objects and Open-Set Relationships
    :house:project
  • FSC: Few-point Shape Completion
  • 3D Paintbrush: Local Stylization of 3D Shapes with Cascaded Score Distillation
    :star:code
    :house:project3D 形状
  • Category-Level Multi-Part Multi-Joint 3D Shape Assembly
  • Neural Point Cloud Diffusion for Disentangled 3D Shape and Appearance Generation
• Stereo Matching
  • Selective-Stereo: Adaptive Frequency Information Selection for Stereo Matching
    :star:code
  • LoS: Local Structure-Guided Stereo Matching
  • Robust Synthetic-to-Real Transfer for Stereo Matching
  • Adaptive Multi-Modal Cross-Entropy Loss for Stereo Matching
  • Neural Markov Random Field for Stereo Matching
    :star:code
  • Reusable Architecture Growth for Continual Stereo Matching
  • MoCha-Stereo: Motif Channel Attention Network for Stereo Matching
    :star:code
    :house:project
  • Learning Intra-view and Cross-view Geometric Knowledge for Stereo Matching
• 表面法线估计
  • Rethinking Inductive Biases for Surface Normal Estimation
    :star:code
• 特征匹配
  • OmniGlue: Generalizable Feature Matching with Foundation Model Guidance
  • RoMa: Robust Dense Feature Matching
  • Efficient LoFTR: Semi-Dense Local Feature Matching with Sparse-Like Speed
    :star:code
• 三维检索
  • KP-RED: Exploiting Semantic Keypoints for Joint 3D Shape Retrieval and Deformation
    :star:code
• 深度补全
  • Flexible Depth Completion for Sparse and Varying Point Densities
  • Improving Depth Completion via Depth Feature Upsampling
  • Test-Time Adaptation for Depth Completion
  • Bilateral Propagation Network for Depth Completion
  • DeCoTR: Enhancing Depth Completion with 2D and 3D Attentions
  • Tri-Perspective View Decomposition for Geometry-Aware Depth Completion
    :star:code
• 深度估计
  • Cross-spectral Gated-RGB Stereo Depth Estimation
  • Mining Supervision for Dynamic Regions in Self-Supervised Monocular Depth Estimation
  • Depth Prompting for Sensor-Agnostic Depth Estimation
  • Atlantis: Enabling Underwater Depth Estimation with Stable Diffusion
    :star:code
  • On the Robustness of Language Guidance for Low-Level Vision Tasks: Findings from Depth Estimation
    :house:project
  • Mind The Edge: Refining Depth Edges in Sparsely-Supervised Monocular Depth Estimation
    :star:code
  • PatchFusion: An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation
  • Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation
    :house:project
  • Elite360D: Towards Efficient 360 Depth Estimation via Semantic- and Distance-Aware Bi-Projection Fusion
  • ECoDepth: Effective Conditioning of Diffusion Models for Monocular Depth Estimation
    :star:code
  • From-Ground-To-Objects: Coarse-to-Fine Self-supervised Monocular Depth Estimation of Dynamic Objects with Ground Contact Prior
  • UniDepth: Universal Monocular Metric Depth Estimation
    :star:code
  • WorDepth: Variational Language Prior for Monocular Depth Estimation
  • SPIDeRS: Structured Polarization for Invisible Depth and Reflectance Sensing
  • Snapshot Lidar: Fourier Embedding of Amplitude and Phase for Single-Image Depth Reconstruction
• 全景定位
  • Fully Geometric Panoramic Localization
    :star:code
• 3D关键点检测
  • Back to 3D: Few-Shot 3D Keypoint Detection with Back-Projected 2D Features
    :house:project
• 布局重建
  • Seg2Reg: Differentiable 2D Segmentation to 1D Regression Rendering for 360 Room Layout Reconstruction
  • No More Ambiguity in 360deg Room Layout via Bi-Layout Estimation
    :star:code
• CAD 重建
  • SfmCAD: Unsupervised CAD Reconstruction by Learning Sketch-based Feature Modeling Operations无监督 CAD 重建
• 形状匹配
  • Hybrid Functional Maps for Crease-Aware Non-Isometric Shape Matching
• 3DGS
  • COLMAP-Free 3D Gaussian Splatting
    :star:code
    :house:project
  • Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields
  • GS-IR: 3D Gaussian Splatting for Inverse Rendering
  • FreGS: 3D Gaussian Splatting with Progressive Frequency Regularization
    :house:project
  • Scaffold-GS: Structured 3D Gaussians for View-Adaptive Rendering
    :house:project
  • GaussianDreamer: Fast Generation from Text to 3D Gaussians by Bridging 2D and 3D Diffusion Models
    :house:project
  • Mip-Splatting: Alias-free 3D Gaussian Splatting
    :star:code
    :house:project
  • CoGS: Controllable Gaussian Splatting
    :star:code
    :house:project
  • LangSplat: 3D Language Gaussian Splatting
    :star:code
    :house:project
  • Compact 3D Gaussian Representation for Radiance Field
    :house:project
  • 3DGStream: On-the-Fly Training of 3D Gaussians for Efficient Streaming of Photo-Realistic Free-Viewpoint Videos
    :house:project
  • [HUGS: Human Gaussian Splatting]
  • HUGS: Human Gaussian Splats
    :star:code
    :house:project
  • Multi-Scale 3D Gaussian Splatting for Anti-Aliased Rendering3DGS
  • GaussianShader: 3D Gaussian Splatting with Shading Functions for Reflective Surfaces
• 场景重建
  • Gated Fields: Learning Scene Reconstruction from Gated Videos
  • Guess The Unseen: Dynamic 3D Scene Reconstruction from Partial 2D Glimpses
  • SuperPrimitive: Scene Reconstruction at a Primitive Level
    :house:project
  • Total-Decom: Decomposed 3D Scene Reconstruction with Minimal Interaction
    :star:code
  • Behind the Veil: Enhanced Indoor 3D Scene Reconstruction with Occluded Surfaces Completion
  • OmniSDF: Scene Reconstruction using Omnidirectional Signed Distance Functions and Adaptive Binoctrees
  • VastGaussian: Vast 3D Gaussians for Large Scene Reconstruction
    :star:code
  • Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction
    :star:code
    :house:project
    :thumbsup:CVPR 2024满分论文：浙大提出基于可变形三维高斯的高质量单目动态重建新方法
  • Polarization Wavefront Lidar: Learning Large Scene Reconstruction from Polarized Wavefronts
    :house:project
• 3D 场景合成
  • GraphDreamer: Compositional 3D Scene Synthesis from Scene Graphs
    :house:project
  • DiffInDScene: Diffusion-based High-Quality 3D Indoor Scene Generation
    :star:code
  • BerfScene: Bev-conditioned Equivariant Radiance Fields for Infinite 3D Scene Generation
    :house:project3D 场景生成
  • Sat2Scene: 3D Urban Scene Generation from Satellite Images with Diffusion
  • 文本驱动的 3D 场景生成
    • 3D-SceneDreamer: Text-Driven 3D-Consistent Scene Generation
    • Towards Text-guided 3D Scene Composition
      :house:project3D 场景合成
• 3D 场景图
  • SG-PGM: Partial Graph Matching Network with Semantic Geometric Fusion for 3D Scene Graph Alignment and Its Downstream Tasks
• 3D 场景编辑
  • GaussianEditor:Editing 3D Gaussians Delicately with Text Instructions
    :house:project
  • Customize your NeRF: Adaptive Source Driven 3D Scene Editing via Local-Global Iterative Training
    :thumbsup:文本或图像提示精准编辑3D场景，美图&信工所&北航&中大联合提出3D编辑方法CustomNeRF
  • PhyScene: Physically Interactable 3D Scene Synthesis for Embodied AI
    :star:code
  • Neural 3D Strokes: Creating Stylized 3D Scenes with Vectorized 3D Strokes
    :house:project3D 场景
  • PAPR in Motion: Seamless Point-level 3D Scene Interpolation3D 场景插值
  • ConsistDreamer: 3D-Consistent 2D Diffusion for High-Fidelity Scene Editing
• 语义匹配
  • SD4Match: Learning to Prompt Stable Diffusion Model for Semantic Matching
    :star:code
    :house:project
• 室内照明估计
  • LightOctree: Lightweight 3D Spatially-Coherent Indoor Lighting Estimation
• 三维服装生成
  • Design2Cloth: 3D Cloth Generation from 2D Masks
    :house:project
• 3D 形状匹配
  • SpiderMatch: 3D Shape Matching with Global Optimality and Geometric Consistency

10.Medical Image Progress(医学影响处理)

• Brain Decodes Deep Nets
• Feature Re-Embedding: Towards Foundation Model-Level Performance in Computational Pathology
  :star:code
• MLIP: Enhancing Medical Visual Representation with Divergence Encoder and Knowledge-guided Contrastive Learning
• Seeing Unseen: Discover Novel Biomedical Concepts via Geometry-Constrained Probabilistic Modeling
• Continual Self-supervised Learning: Towards Universal Multi-modal Medical Data Representation Learning
  :star:code
• MindBridge: A Cross-Subject Brain Decoding Framework
  :star:code
  :star:code
• MedM2G: Unifying Medical Multi-Modal Generation via Cross-Guided Diffusion with Visual Invariant
• Data-Efficient Unsupervised Interpolation Without Any Intermediate Frame for 4D Medical Images
  :star:code
• PairAug: What Can Augmented Image-Text Pairs Do for Radiology?
  :star:code
• Tumor Micro-environment Interactions Guided Graph Learning for Survival Analysis of Human Cancers from Whole-slide Pathological Images
• C^2RV: Cross-Regional and Cross-View Learning for Sparse-View CBCT Reconstruction
  :star:code
• VoCo: A Simple-yet-Effective Volume Contrastive Learning Framework for 3D Medical Image Analysis
• Think Twice Before Selection: Federated Evidential Active Learning for Medical Image Analysis with Domain Shifts
• CT
  • QN-Mixer: A Quasi-Newton MLP-Mixer Model for Sparse-View CT Reconstruction
    :star:code
  • Bootstrapping Chest CT Image Understanding by Distilling Knowledge from X-ray Expert Models
• 切片分类
  • Generalizable Whole Slide Image Classification with Fine-Grained Visual-Semantic Interaction
  • Dynamic Graph Representation with Knowledge-aware Attention for Histopathology Whole Slide Image Analysis
    :star:code
  • Dynamic Policy-Driven Adaptive Multi-Instance Learning for Whole Slide Image Classification
    :house:project
  • ViLa-MIL: Dual-scale Vision-Language Multiple Instance Learning for Whole Slide Image Classification
• 肿瘤合成
  • Towards Generalizable Tumor Synthesis
• 病理检测
  • Decomposing Disease Descriptions for Enhanced Pathology Detection: A Multi-Aspect Vision-Language Pre-training Framework
    :star:code
• 基因检测
  • Accurate Spatial Gene Expression Prediction by integrating Multi-resolution features
    :star:code基因预测
• 癌症检测
  • FocusMAE: Gallbladder Cancer Detection from Ultrasound Videos with Focused Masked Autoencoders
    :star:code
• 医学图像配准
  • Correlation-aware Coarse-to-fine MLPs for Deformable Medical Image Registration入围最佳论文
  • Modality-Agnostic Structural Image Representation Learning for Deformable Multi-Modality Medical Image Registration
• 医学图像分类
  • Systematic Comparison of Semi-supervised and Self-supervised Learning for Medical Image Classification
• 医学图像分割
  • Training Like a Medical Resident: Context-Prior Learning Toward Universal Medical Image Segmentation
  • One-Prompt to Segment All Medical Images
  • Diversified and Personalized Multi-rater Medical Image Segmentation
    :star:code
  • Unleashing the Potential of SAM for Medical Adaptation via Hierarchical Decoding
    :star:code
  • Versatile Medical Image Segmentation Learned from Multi-Source Datasets via Model Self-Disambiguation
  • Adaptive Bidirectional Displacement for Semi-Supervised Medical Image Segmentation
  • MAPSeg: Unified Unsupervised Domain Adaptation for Heterogeneous Medical Image Segmentation Based on 3D Masked Autoencoding and Pseudo-Labeling
  • EMCAD: Efficient Multi-scale Convolutional Attention Decoding for Medical Image Segmentation
  • Tyche: Stochastic In-Context Learning for Medical Image Segmentation
  • Modality-agnostic Domain Generalizable Medical Image Segmentation by Multi-Frequency in Multi-Scale Attention
    :house:project
  • Clustering Propagation for Universal Medical Image Segmentation
  • Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling无监督语义分割
  • MemSAM: Taming Segment Anything Model for Echocardiography Video Segmentation
    :star:code超声心动图视频分割
  • Bi-level Learning of Task-Specific Decoders for Joint Registration and One-Shot Medical Image Segmentation
  • Constructing and Exploring Intermediate Domains in Mixed Domain Semi-supervised Medical Image Segmentation
    :star:code
  • Incremental Nuclei Segmentation from Histopathological Images via Future-class Awareness and Compatibility-inspired Distillation
    :star:code细胞核分割
  • PH-Net: Semi-Supervised Breast Lesion Segmentation via Patch-wise Hardness
    :star:code半监督乳腺病变分割
  • PrPSeg: Universal Proposition Learning for Panoramic Renal Pathology Segmentation全景肾脏病理分割
  • Each Test Image Deserves A Specific Prompt: Continual Test-Time Adaptation for 2D Medical Image Segmentation
    :star:code
• X-ray
  • Intraoperative 2D/3D Image Registration via Differentiable X-ray Rendering
    :star:code
• MRI
  • Progressive Divide-and-Conquer via Subsampling Decomposition for Accelerated MRI
  • Rethinking Diffusion Model for Multi-Contrast MRI Super-Resolution
  • Fully Convolutional Slice-to-Volume Reconstruction for Single-Stack MRI
    :star:code
• 异常检测
  • Adapting Visual-Language Models for Generalizable Anomaly Detection in Medical Images
    :star:code
• 脑活动
  • Psychometry: An Omnifit Model for Image Reconstruction from Human Brain Activity
• 生存预测
  • Modeling Dense Multimodal Interactions Between Biological Pathways and Histology for Survival Prediction
    :star:code
• 计算病理学
  • Transcriptomics-guided Slide Representation Learning in Computational Pathology
  • XFibrosis: Explicit Vessel-Fiber Modeling for Fibrosis Staging from Liver Pathology Images
• 组织病理学
  • SI-MIL: Taming Deep MIL for Self-Interpretability in Gigapixel Histopathology
  • CPLIP: Zero-Shot Learning for Histopathology with Comprehensive Vision-Language Alignment
  • Prompting Vision Foundation Models for Pathology Image Analysis
  • Quilt-LLaVA: Visual Instruction Tuning by Extracting Localized Narratives from Open-Source Histopathology Videos
    :house:project
• 医学超分辨率
  • CycleINR: Cycle Implicit Neural Representation for Arbitrary-Scale Volumetric Super-Resolution of Medical Data
• 3D医学影像
  • ToNNO: Tomographic Reconstruction of a Neural Network's Output for Weakly Supervised Segmentation of 3D Medical Images
• 放射学报告生成
  • Instance-level Expert Knowledge and Aggregate Discriminative Attention for Radiology Report Generation
• 放射学报告检索
  • AHIVE: Anatomy-aware Hierarchical Vision Encoding for Interactive Radiology Report Retrieval
• 医学基础模型
  • Low-Rank Knowledge Decomposition for Medical Foundation Models
• 肿瘤分割
  • ZePT: Zero-Shot Pan-Tumor Segmentation via Query-Disentangling and Self-Prompting
• 基因表达预测
  • Accurate Spatial Gene Expression Prediction by Integrating Multi-Resolution Features

9.Face(人脸)

• Unsupervised Gaze Representation Learning from Multi-view Face Images
• ToonerGAN: Reinforcing GANs for Obfuscating Automated Facial Indexing
• PairDETR : Joint Detection and Association of Human Bodies and Faces
• Neural Implicit Morphing of Face Images
  :house:project
• SimAC: A Simple Anti-Customization Method for Protecting Face Privacy against Text-to-Image Synthesis of Diffusion Models
• Anatomically Constrained Implicit Face Models
• Face2Diffusion for Fast and Editable Face Personalization
  :star:code
  :star:code
• Diffusion-based Adversarial Makeup Transfer for Facial Privacy Protection
  :star:code
• Self-Supervised Facial Representation Learning with Facial Region Awareness
• Facial Identity Anonymization via Intrinsic and Extrinsic Attention Distraction
• VOODOO 3D: Volumetric Portrait Disentanglement For One-Shot 3D Head Reenactment
• 人脸编辑
  • StrokeFaceNeRF: Stroke-based Facial Appearance Editing in Neural Radiance Field
  • In-N-Out: Faithful 3D GAN Inversion with Volumetric Decomposition for Face Editing
    :house:project
• 人脸表情
  • Learning Adaptive Spatial Coherent Correlations for Speech-Preserving Facial Expression Manipulation
  • 3D Facial Expressions through Analysis-by-Neural-Synthesis
    :house:project
• 人脸识别
  • OpticalDR: A Deep Optical Imaging Model for Privacy-Protective Depression Recognition(抑郁症识别)
  • Privacy-Preserving Face Recognition Using Trainable Feature Subtraction
    :star:code
  • KeyPoint Relative Position Encoding for Face Recognition
  • LAFS: Landmark-based Facial Self-supervised Learning for Face Recognition
  • Validating Privacy-Preserving Face Recognition under a Minimum Assumption
• 人脸合成
  • Deformable One-shot Face Stylization via DINO Semantic Guidance
    :star:code
  • Towards a Simultaneous and Granular Identity-Expression Control in Personalized Face Generation
  • DreamSalon: A Staged Diffusion Framework for Preserving Identity-Context in Editable Face Generation
  • LeGO: Leveraging a Surface Deformation Network for Animatable Stylized Face Generation with One Example
    :star:code
    :house:project
  • Text-Guided 3D Face Synthesis - From Generation to Editing
  • Text-conditional Attribute Alignment across Latent Spaces for 3D Controllable Face Image Synthesis
  • UV-IDM: Identity-Conditioned Latent Diffusion Model for Face UV-Texture Generation
  • Diffusion-driven GAN Inversion for Multi-Modal Face Image Generation
• 人脸重建
  • High-Quality Facial Geometry and Appearance Capture at Home
    :star:code
    :house:project
  • Monocular Identity-Conditioned Facial Reflectance Reconstruction
    :star:code
    :thumbsup:三维数字人重建、编辑与驱动
  • 3D Face Reconstruction with the Geometric Guidance of Facial Part Segmentation
    :star:code
  • 3D-Aware Face Editing via Warping-Guided Latent Direction Learning
    :house:project
    :thumbsup:三维数字人重建、编辑与驱动
• 人脸修饰
  • VRetouchEr: Learning Cross-frame Feature Interdependence with Imperfection Flow for Face Retouching in Videos
• 人脸重现
  • Pose Adapted Shape Learning for Large-Pose Face Reenactment
  • FSRT: Facial Scene Representation Transformer for Face Reenactment from Factorized Appearance Head-pose and Facial Expression Features
• 人脸恢复
  • Learning Degradation-unaware Representation with Prior-based Latent Transformations for Blind Face Restoration
  • PFStorer: Personalized Face Restoration and Super-Resolution
  • WaveFace: Authentic Face Restoration with Efficient Frequency Recovery
• 人脸去识别
  • Privacy-preserving Optics for Enhancing Protection in Face De-identification
    :house:project
• 人脸化妆
  • Makeup Prior Models for 3D Facial Makeup Estimation and Applications
    :star:code
• 人脸关键点
  • Generalizable Face Landmarking Guided by Conditional Face Warping
    :star:code
    :house:project
  • FaceLift: Semi-supervised 3D Facial Landmark Localization
• 人脸属性分类
  • Distributionally Generative Augmentation for Fair Facial Attribute Classification
    :star:code
• 人脸活体检测
  • One-Class Face Anti-spoofing via Spoof Cue Map-Guided Feature Learning
  • Rethinking Generalizable Face Anti-spoofing via Hierarchical Prototype-guided Distribution Refinement in Hyperbolic Space
  • CFPL-FAS: Class Free Prompt Learning for Generalizable Face Anti-spoofing
  • Suppress and Rebalance: Towards Generalized Multi-Modal Face Anti-Spoofing
    :star:code
  • Gradient Alignment for Cross-Domain Face Anti-Spoofing
    :star:code
  • Test-Time Domain Generalization for Face Anti-Spoofing
  • Gradient Alignment for Cross-domain Face Anti-Spoofing
    :star:code
• 人脸动作单元
  • Multi-scale Dynamic and Hierarchical Relationship Modeling for Facial Action Units Recognition
  • AUEditNet: Dual-Branch Facial Action Unit Intensity Manipulation with Implicit Disentanglement
• 人脸图像质量
  • CLIB-FIQA: Face Image Quality Assessment with Confidence Calibration
  • DSL-FIQA: Assessing Facial Image Quality via Dual-Set Degradation Learning and Landmark-Guided Transformer
• 肖像编辑
  • Control4D: Efficient 4D Portrait Editing with Text
    :house:project
• 头发重建
  • Dr.Hair: Reconstructing Scalp-Connected Hair Strands without Pre-Training via Differentiable Rendering of Line Segments
• 三维人脸
  • 3D Face Tracking from 2D Video through Iterative Dense UV to Image Flow
  • FaceCom: Towards High-fidelity 3D Facial Shape Completion via Optimization and Inpainting Guidance
  • Probabilistic Speech-Driven 3D Facial Motion Synthesis: New Benchmarks Methods and Applications
• 4D 头像合成
  • Portrait4D: Learning One-Shot 4D Head Avatar Synthesis using Synthetic Data
    :star:code
    :house:project
• 头像重建
  • HAVE-FUN: Human Avatar Reconstruction from Few-Shot Unconstrained Images
    :house:project
  • MoSAR: Monocular Semi-Supervised Model for Avatar Reconstruction using Differentiable Shading
    :house:project
  • Morphable Diffusion: 3D-Consistent Diffusion for Single-image Avatar Creation
    :house:project
• 说话头合成
  • Learning Dynamic Tetrahedra for High-Quality Talking Head Synthesis
  • Faces that Speak: Jointly Synthesising Talking Face and Speech from Text
  • CustomListener: Text-guided Responsive Interaction for User-friendly Listening Head Generation
  • SyncTalk: The Devil is in the Synchronization for Talking Head Synthesis
    :star:code
    :house:project
  • FaceChain-ImagineID: Freely Crafting High-Fidelity Diverse Talking Faces from Disentangled Audio
    :star:code
  • FaceTalk: Audio-Driven Motion Diffusion for Neural Parametric Head Models
    :star:code
    :house:project
  • FlowVQTalker: High-Quality Emotional Talking Face Generation through Normalizing Flow and Quantization
• 防御人脸编辑滥用
  • IDGuard: Robust General Identity-centric POI Proactive Defense Against Face Editing Abuse
• 3D 头像
  • 3DToonify: Creating Your High-Fidelity 3D Stylized Avatar Easily from 2D Portrait Images
  • DreamAvatar: Text-and-Shape Guided 3D Human Avatar Generation via Diffusion Models
    :house:project3
• 化妆迁移
  • Content-Style Decoupling for Unsupervised Makeup Transfer without Generating Pseudo Ground Truth
• 人脸重识别
  • Privacy-Preserving Optics for Enhancing Protection in Face De-Identification
• 年龄估计
  • A Call to Reflect on Evaluation Practices for Age Estimation: Comparative Analysis of the State-of-the-Art and a Unified Benchmark
• 情绪识别
  • Robust Emotion Recognition in Context Debiasing

8.GAN/Image Synthesis(图像生成)

• L-MAGIC: Language Model Assisted Generation of Images with Coherence
• CapsFusion: Rethinking Image-Text Data at Scale
• C3Net: Compound Conditioned ControlNet for Multimodal Content Generation
• Scaling Laws of Synthetic Images for Model Training ... for Now
• An edit friendly ddpm noise space: inversion and manipulations
  :star:code
  :house:project
• CoDi-2: In-Context Interleaved and Interactive Any-to-Any Generation
  :star:code
  :house:project
• CapHuman: Capture Your Moments in Parallel Universes
  :star:code
  :house:project
• Text-Conditioned Generative Model of 3D Strand-based Human Hairstyles
  :house:project
• IMPRINT: Generative Object Compositing by Learning Identity-Preserving Representation
• TexTile: A Differentiable Metric for Texture Tileability
  :house:project
• SD-DiT: Unleashing the Power of Self-supervised Discrimination in Diffusion Transformer
• PhotoMaker: Customizing Realistic Human Photos via Stacked ID Embedding
  :star:code
  :house:project
• MobileCLIP: Fast Image-Text Models through Multi-Modal Reinforced Training
  :star:code
• Text-Image Alignment for Diffusion-Based Perception
• AEROBLADE: Training-Free Detection of Latent Diffusion Images Using Autoencoder Reconstruction Error
• FaceChain-SuDe: Building Derived Class to Inherit Category Attributes for One-shot Subject-Driven Generation
  :star:code
• It's All About Your Sketch: Democratising Sketch Control in Diffusion Models
  :star:code
• Codebook Transfer with Part-of-Speech for Vector-Quantized Image Modeling
• ProMark: Proactive Diffusion Watermarking for Causal Attribution
• DetDiffusion: Synergizing Generative and Perceptive Models for Enhanced Data Generation and Perception
• GAN
  • StyLitGAN: Image-Based Relighting via Latent Control
    :star:code
    :house:project
  • StyleCineGAN: Landscape Cinemagraph Generation using a Pre-trained StyleGAN
  • What You See is What You GAN: Rendering Every Pixel for High-Fidelity Geometry in 3D GANs
    :house:project
  • Diversity-aware Channel Pruning for StyleGAN Compression
    :star:code
  • Adversarial Score Distillation: When score distillation meets GAN
    :star:code
    :house:project
• 扩散
  • Fixed Point Diffusion Models
    :house:project
  • Diffusion Models Without Attention
  • Image Neural Field Diffusion Models
  • Functional Diffusion
    :house:project
  • Smooth Diffusion: Crafting Smooth Latent Spaces in Diffusion Models
  • Learned Representation-Guided Diffusion Models for Large-Image Generation
  • ACT-Diffusion: Efficient Adversarial Consistency Training for One-step Diffusion Models
  • Using Human Feedback to Fine-tune Diffusion Models without Any Reward Model
  • LightIt: Illumination Modeling and Control for Diffusion Models
  • Towards More Accurate Diffusion Model Acceleration with A Timestep Tuner
  • MMA-Diffusion: MultiModal Attack on Diffusion Models
  • CommonCanvas: Open Diffusion Models Trained on Creative-Commons Images
  • Can Protective Perturbation Safeguard Personal Data from Being Exploited by Stable Diffusion?
  • Self-correcting LLM-controlled Diffusion Models
  • Attention-Driven Training-Free Efficiency Enhancement of Diffusion Models
  • SODA: Bottleneck Diffusion Models for Representation Learning
  • PRDP: Proximal Reward Difference Prediction for Large-Scale Reward Finetuning of Diffusion Models
  • Don't drop your samples! Coherence-aware training benefits Conditional diffusion
    :house:project
  • Improving Training Efficiency of Diffusion Models via Multi-Stage Framework and Tailored Multi-Decoder Architecture
  • DiffLoc: Diffusion Model for Outdoor LiDAR Localization
    :thumbsup:摘要
  • EasyDrag: Efficient Point-based Manipulation on Diffusion Models
  • Distilling ODE Solvers of Diffusion Models into Smaller Steps
  • Cache Me if You Can: Accelerating Diffusion Models through Block Caching
    :house:project
  • Beyond Textual Constraints: Learning Novel Diffusion Conditions with Fewer Examples
  • AAMDM: Accelerated Auto-regressive Motion Diffusion Model
  • DeepCache: Accelerating Diffusion Models for Free
    :house:project
  • Diffusion Model Alignment Using Direct Preference Optimization
  • Perturbing Attention Gives You More Bang for the Buck: Subtle Imaging Perturbations That Efficiently Fool Customized Diffusion Models
  • Analyzing and Improving the Training Dynamics of Diffusion Models
  • Residual Learning in Diffusion Models
  • FreeU: Free Lunch in Diffusion U-Net
    :star:code
    :house:project
  • VideoCrafter2: Overcoming Data Limitations for High-Quality Video Diffusion Models
    :star:code
    :house:project
  • Diff-BGM: A Diffusion Model for Video Background Music Generation视频背景音乐生成的扩散模型
  • Your Student is Better Than Expected: Adaptive Teacher-Student Collaboration for Text-Conditional Diffusion Models
  • Shadow Generation for Composite Image Using Diffusion Model
    :star:code
  • Alchemist: Parametric Control of Material Properties with Diffusion Models
  • Orthogonal Adaptation for Modular Customization of Diffusion Models
    :house:project扩散模型
  • Observation-Guided Diffusion Probabilistic Models
    :star:code
  • TFMQ-DM: Temporal Feature Maintenance Quantization for Diffusion Models
    :star:code
  • Visual Anagrams: Generating Multi-View Optical Illusions with Diffusion Models
  • SPAD: Spatially Aware Multi-View Diffusers
    :house:project
  • Structure-Guided Adversarial Training of Diffusion Models
  • One-step Diffusion with Distribution Matching Distillation
    :house:project
  • Rethinking the Spatial Inconsistency in Classifier-Free Diffusion Guidance
    :star:code
  • Intelligent Grimm -- Open-ended Visual Storytelling via Latent Diffusion Models
    :star:code
    :house:project
  • X-Adapter: Adding Universal Compatibility of Plugins for Upgraded Diffusion Model
    :house:project
  • Readout Guidance: Learning Control from Diffusion Features
    :house:project
  • PointInfinity: Resolution-Invariant Point Diffusion Models
    :house:project
  • Unsupervised Keypoints from Pretrained Diffusion Models
    :star:code
  • Amodal Completion via Progressive Mixed Context Diffusion
    :star:code
    :house:project
  • SkillDiffuser: Interpretable Hierarchical Planning via Skill Abstractions in Diffusion-Based Task Execution
    :house:project
  • DREAM: Diffusion Rectification and Estimation-Adaptive Models
  • Towards Memorization-Free Diffusion Models
  • Efficient Dataset Distillation via Minimax Diffusion
    :star:code
  • MatFuse: Controllable Material Generation with Diffusion Models
    :star:code
    :house:project
  • Accelerating Diffusion Sampling with Optimized Time Steps
  • Boosting Diffusion Models with Moving Average Sampling in Frequency Domain
  • One-dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applications
    :star:code
    :house:project
  • Balancing Act: Distribution-Guided Debiasing in Diffusion Models
    :star:code
  • Shadow Generation for Composite Image Using Diffusion model
    :star:code
  • MACE: Mass Concept Erasure in Diffusion Models
    :star:code
  • DistriFusion: Distributed Parallel Inference for High-Resolution Diffusion Models
    :house:project
    :house:project
    :star:code
  • Tackling the Singularities at the Endpoints of Time Intervals in Diffusion Models
    :star:code
    :house:project
  • DEADiff: An Efficient Stylization Diffusion Model with Disentangled Representations
    :star:code
    :house:project
  • SVGDreamer: Text Guided SVG Generation with Diffusion Model
    :star:code
    :house:project
    :thumbsup:SVGDreamer: 北航&港大发布全新文本引导的矢量图形可微渲染方法
  • Relation Rectification in Diffusion Model
    :star:code
    :house:project
• 图像合成/生成
  • 图像合成
    • One-Shot Structure-Aware Stylized Image Synthesis
    • AnyScene: Customized Image Synthesis with Composited Foreground
    • Improving Subject-Driven Image Synthesis with Subject-Agnostic Guidance
    • ViewFusion: Towards Multi-View Consistency via Interpolated Denoising
      :star:code
    • PLACE: Adaptive Layout-Semantic Fusion for Semantic Image Synthesis
    • Diff-Plugin: Revitalizing Details for Diffusion-based Low-level Tasks
    • Rethinking the Objectives of Vector-Quantized Tokenizers for Image Synthesis
    • Unmixing Before Fusion: A Generalized Paradigm for Multi-Source-based Hyperspectral Image Synthesis
    • [Unlocking Pretrained Image Backbones for Semantic Image Synthesis]
    • Unlocking Pre-trained Image Backbones for Semantic Image Synthesis
    • Revisiting Non-Autoregressive Transformers for Efficient Image Synthesis
    • 场景-文本图像合成
      • TextNeRF: A Novel Scene-Text Image Synthesis Method based on Neural Radiance Fields
  • 图像生成
    • [ElasticDiffusion: Training-free Arbitrary Size Image Generation]
    • ElasticDiffusion: Training-free Arbitrary Size Image Generation through Global-Local Content Separation
      :house:project
    • SCoFT: Self-Contrastive Fine-Tuning for Equitable Image Generation
    • AnyDoor: Zero-shot Object-level Image Customization
      :house:project
    • Taming Stable Diffusion for Text to 360 Panorama Image Generation
      :star:code
      :star:code
    • Active Open-Vocabulary Recognition: Let Intelligent Moving Mitigate CLIP Limitations
    • Generative Image Dynamics
      :house:project
    • Clockwork Diffusion: Efficient Generation With Model-Step Distillation
    • UniGS: Unified Representation for Image Generation and Segmentation
      :star:code图像生成
    • Exact Fusion via Feature Distribution Matching for Few-shot Image Generation
    • FreeCustom: Tuning-Free Customized Image Generation for Multi-Concept Composition
      :star:code
      :house:project
    • Adversarial Text to Continuous Image Generation
    • Style Aligned Image Generation via Shared Attention
      :house:project
    • CoDi: Conditional Diffusion Distillation for Higher-Fidelity and Faster Image Generation
    • Instruct-Imagen: Image Generation with Multi-modal Instruction
    • InstanceDiffusion: Instance-level Control for Image Generation
      :star:code
      :house:project
    • DemoFusion: Democratising High-Resolution Image Generation With No $$$
      :star:code
      :house:project
    • ViewDiff: 3D-Consistent Image Generation with Text-to-Image Models
      :star:code
      :house:project
      :star:code
    • When StyleGAN Meets Stable Diffusion:a W+ Adapter for Personalized Image Generation
      :star:code
      :house:project
    • Correcting Diffusion Generation through Resampling
      :star:code
    • Arbitrary-Scale Image Generation and Upsampling using Latent Diffusion Model and Implicit Neural Decoder
    • Condition-Aware Neural Network for Controlled Image Generation
    • A Unified and Interpretable Emotion Representation and Expression Generation
      :star:code
    • Rethinking FID: Towards a Better Evaluation Metric for Image Generation
    • 主题驱动的图像生成
      • SSR-Encoder: Encoding Selective Subject Representation for Subject-Driven Generation
        :star:code
        :house:project
  • 文本-图像
    • Zero-Painter: Training-Free Layout Control for Text-to-Image Synthesis
    • Learning Multi-Dimensional Human Preference for Text-to-Image Generation
    • Customization Assistant for Text-to-Image Generation
    • TokenCompose: Text-to-Image Diffusion with Token-level Supervision
    • FreeControl: Training-Free Spatial Control of Any Text-to-Image Diffusion Model with Any Condition
    • Personalized Residuals for Concept-Driven Text-to-Image Generation
      :house:project
    • Rich Human Feedback for Text-to-Image Generation
    • MarkovGen: Structured Prediction for Efficient Text-to-Image Generation
    • Intriguing Properties of Diffusion Models: An Empirical Study of the Natural Attack Capability in Text-to-Image Generative Models
    • Customization Assistant for Text-to-image Generation
    • SwiftBrush: One-Step Text-to-Image Diffusion Model with Variational Score Distillation
    • JeDi: Joint-Image Diffusion Models for Finetuning-Free Personalized Text-to-Image Generation
    • MIGC: Multi-Instance Generation Controller for Text-to-Image Synthesis
      :house:project
    • Check, Locate, Rectify: A Training-Free Layout Calibration System for Text-to-Image Generation
      :house:project
    • Predicated Diffusion: Predicate Logic-Based Attention Guidance for Text-to-Image Diffusion Models
    • DreamMatcher: Appearance Matching Self-Attention for Semantically-Consistent Text-to-Image Personalization
      :house:project
    • UFOGen: You Forward Once Large Scale Text-to-Image Generation via Diffusion GANs
    • Prompting Hard or Hardly Prompting: Prompt Inversion for Text-to-Image Diffusion Models
    • Countering Personalized Text-to-Image Generation with Influence Watermarks
    • Ranni: Taming Text-to-Image Diffusion for Accurate Instruction Following
      :house:project
    • Tailored Visions: Enhancing Text-to-Image Generation with Personalized Prompt Rewriting
      :star:code
    • InstantBooth: Personalized Text-to-Image Generation without Test-Time Finetuning
    • FakeInversion: Learning to Detect Images from Unseen Text-to-Image Models by Inverting Stable Diffusion
    • Learning Disentangled Identifiers for Action-Customized Text-to-Image Generation
      :house:project
    • LeftRefill: Filling Right Canvas based on Left Reference through Generalized Text-to-Image Diffusion Model
      :star:code
      :house:project
    • HyperDreamBooth: HyperNetworks for Fast Personalization of Text-to-Image Models
      :house:project
    • PIA: Your Personalized Image Animator via Plug-and-Play Modules in Text-to-Image Models
      :house:project
    • On the Scalability of Diffusion-based Text-to-Image Generation
    • Self-Discovering Interpretable Diffusion Latent Directions for Responsible Text-to-Image Generation
      :house:project
    • EmoGen: Emotional Image Content Generation with Text-to-Image Diffusion Models
      :star:code
    • Grounded Text-to-Image Synthesis with Attention Refocusing
      :house:project
    • OpenBias: Open-set Bias Detection in Text-to-Image Generative Models
      :star:code
    • Prompt-Free Diffusion: Taking “Text” out of Text-to-Image Diffusion Models
      :star:code
    • CONFORM: Contrast is All You Need for High-Fidelity Text-to-Image Diffusion Models
    • InitNO: Boosting Text-to-Image Diffusion Models via Initial Noise Optimization
      :star:code
    • Concept Weaver: Enabling Multi-Concept Fusion in Text-to-Image Models
    • Cross Initialization for Face Personalization of Text-to-Image Models文本到图像Cross Initialization for Personalized Text-to-Image Generation
    • CosmicMan: A Text-to-Image Foundation Model for Humans
      :star:code
    • Dynamic Prompt Optimizing for Text-to-Image Generation
      :star:code
    • WOUAF: Weight Modulation for User Attribution and Fingerprinting in Text-to-Image Diffusion Models
    • Attention Calibration for Disentangled Text-to-Image Personalization
    • RealCustom: Narrowing Real Text Word for Real-Time Open-Domain Text-to-Image Customization
      :star:code
    • InteractDiffusion: Interaction Control in Text-to-Image Diffusion Models
      :star:code
      :house:project
    • Learning Continuous 3D Words for Text-to-Image Generation
      :star:code
      :house:project
    • NoiseCollage: A Layout-Aware Text-to-Image Diffusion Model Based on Noise Cropping and Merging
      :star:code
    • HanDiffuser: Text-to-Image Generation With Realistic Hand Appearances
      :house:project
    • Discriminative Probing and Tuning for Text-to-Image Generation
      :star:code
      :house:project
    • Selectively Informative Description can Reduce Undesired Embedding Entanglements in Text-to-Image Personalization
    • ECLIPSE: A Resource-Efficient Text-to-Image Prior for Image Generations
      :star:code
      :house:project
    • FlashEval: Towards Fast and Accurate Evaluation of Text-to-image Diffusion Generative Models
      :star:code
    • MetaCloak: Preventing Unauthorized Subject-driven Text-to-image Diffusion-based Synthesis via Meta-learning
  • 主题-图像
    • High-fidelity Person-centric Subject-to-Image Synthesis
      :star:code
    • [High Fidelity Person-centric Subject-to-Image Synthesis]
• 视频合成/生成
  • 视频生成
    • InstructVideo: Instructing Video Diffusion Models with Human Feedback
      :star:code
      :house:project
    • Make Pixels Dance: High-Dynamic Video Generation
    • GenTron: Diffusion Transformers for Image and Video Generation
    • Panacea: Panoramic and Controllable Video Generation for Autonomous Driving
    • Generative Rendering: Controllable 4D-Guided Video Generation with 2D Diffusion Models
      :house:project
    • DiffPerformer: Iterative Learning of Consistent Latent Guidance for Diffusion-based Human Video Generation
    • VideoBooth: Diffusion-based Video Generation with Image Prompts
      :house:project
    • Hierarchical Patch Diffusion Models for High-Resolution Video Generation
      :star:code
    • On the Content Bias in Fréchet Video Distance
      :star:code
    • 360DVD: Controllable Panorama Video Generation with 360-Degree Video Diffusion Model
    • SimDA: Simple Diffusion Adapter for Efficient Video Generation
      :star:code
      :house:project
    • GenHowTo: Learning to Generate Actions and State Transformations from Instructional Videos视频生成
    • FRESCO: Spatial-Temporal Correspondence for Zero-Shot Video Translation
      :house:project
      :star:code
    • Vlogger: Make Your Dream A Vlog
      :star:code
      :house:project
    • LAMP: Learn A Motion Pattern for Few-Shot Video Generation
      :house:project
    • EvalCrafter: Benchmarking and Evaluating Large Video Generation Models
      :house:project
    • Co-Speech Gesture Video Generation via Motion-Decoupled Diffusion Model
      :star:code
    • BIVDiff: A Training-free Framework for General-Purpose Video Synthesis via Bridging Image and Video Diffusion Models
      :star:code
      :house:project视频合成
    • DreamVideo: Composing Your Dream Videos with Customized Subject and Motion
      :house:project
    • PEEKABOO: Interactive Video Generation via Masked-Diffusion
      :house:project
  • 文本-视频
    • Grid Diffusion Models for Text-to-Video Generation
    • Breathing Life Into Sketches Using Text-to-Video Priors
      :star:code
      :house:project
    • Dysen-VDM: Empowering Dynamics-aware Text-to-Video Diffusion with LLMs
    • TI2V-Zero: Zero-Shot Image Conditioning for Text-to-Video Diffusion Models
      :house:project
    • Animate Anyone: Consistent and Controllable Image-to-Video Synthesis for Character Animation
    • Hierarchical Spatio-temporal Decoupling for Text-to-Video Generation
      :house:project
    • A Recipe for Scaling up Text-to-Video Generation with Text-free Videos
      :house:project
    • TRIP: Temporal Residual Learning with Image Noise Prior for Image-to-Video Diffusion Models
      :star:code
    • Snap Video: Scaled Spatiotemporal Transformers for Text-to-Video Synthesis
      :house:project
    • VMC: Video Motion Customization using Temporal Attention Adaption for Text-to-Video Diffusion Models
      :house:project
    • MicroCinema: A Divide-and-Conquer Approach for Text-to-Video Generation
  • 图像-视频
    • Your Image is My Video: Reshaping the Receptive Field via Image-To-Video Differentiable AutoAugmentation and Fusion
  • 视频-视频
    • Fairy: Fast Parallelized Instruction-Guided Video-to-Video Synthesis
      :house:project
    • FlowVid: Taming Imperfect Optical Flows for Consistent Video-to-Video Synthesis
• 纹理生成/合成
  • 文本-纹理合成
    • Paint-it: Text-to-Texture Synthesis via Deep Convolutional Texture Map Optimization and Physically-Based Rendering
      :house:project
  • 纹理合成
    • SceneTex: High-Quality Texture Synthesis for Indoor Scenes via Diffusion Priors
      :house:project
    • Single Mesh Diffusion Models with Field Latents for Texture Generation
    • TextureDreamer: Image-Guided Texture Synthesis Through Geometry-Aware Diffusion
• 文本-3D
  • DIRECT-3D: Learning Direct Text-to-3D Generation on Massive Noisy 3D Data
    :star:code
    :house:project
  • PI3D: Efficient Text-to-3D Generation with Pseudo-Image Diffusion
  • Text-to-3D using Gaussian Splatting
    :star:code
    :house:project
  • DreamPropeller: Supercharge Text-to-3D Generation with Parallel Sampling
  • RichDreamer: A Generalizable Normal-Depth Diffusion Model for Detail Richness in Text-to-3D
  • Consistent3D: Towards Consistent High-Fidelity Text-to-3D Generation with Deterministic Sampling Prior
  • Sculpt3D: Multi-View Consistent Text-to-3D Generation with Sparse 3D Prior
    :star:code
  • LucidDreamer: Towards High-Fidelity Text-to-3D Generation via Interval Score Matching
    :star:code
  • Direct2.5: Diverse Text-to-3D Generation via Multi-view 2.5D Diffusion
    :house:project
  • Sherpa3D: Boosting High-Fidelity Text-to-3D Generation via Coarse 3D Prior
    :house:project文本到 3D
  • Taming Mode Collapse in Score Distillation for Text-to-3D Generation
    :house:project
  • Text-to-3D Generation with Bidirectional Diffusion using both 2D and 3D priors
    :house:project
  • DreamControl: Control-Based Text-to-3D Generation with 3D Self-Prior
    :star:code
  • VP3D: Unleashing 2D Visual Prompt for Text-to-3D Generation
    :star:code
  • GPT-4V(ision) is a Human-Aligned Evaluator for Text-to-3D Generation
    :star:code
    :house:project
  • Enhancing 3D Fidelity of Text-to-3D using Cross-View Correspondences
  • DiffusionGAN3D: Boosting Text-guided 3D Generation and Domain Adaptation by Combining 3D GANs and Diffusion Priors
    :house:project
  • HyperSDFusion: Bridging Hierarchical Structures in Language and Geometry for Enhanced 3D Text2Shape Generation
  • Diffusion Handles: Enabling 3D Edits for Diffusion Models by Lifting Activations to 3D
    :house:project
  • HarmonyView: Harmonizing Consistency and Diversity in One-Image-to-3D
    :house:project
• 图像-3D
  • Diffusion Time-step Curriculum for One Image to 3D Generation
    :star:code
  • MPOD123: One Image to 3D Content Generation Using Mask-enhanced Progressive Outline-to-Detail Optimization
  • Generative 3D Part Assembly via Part-Whole-Hierarchy Message Passing
• 文本-4D
  • 4D-fy: Text-to-4D Generation Using Hybrid Score Distillation Sampling
    :house:project
  • Align Your Gaussians: Text-to-4D with Dynamic 3D Gaussians and Composed Diffusion Models
    :house:project
• 3D生成
  • DreamComposer: Controllable 3D Object Generation via Multi-View Conditions
    :house:project
  • XCube (X3): Large-Scale 3D Generative Modeling using Sparse Voxel Hierarchies
    :house:project
  • CAD: Photorealistic 3D Generation via Adversarial Distillation
    :star:code
    :house:project
  • Paint3D: Paint Anything 3D with Lighting-Less Texture Diffusion Models
    :star:code3D 内容
  • Interactive3D: Create What You Want by Interactive 3D Generation
• 语义场景生成
  • SemCity: Semantic Scene Generation with Triplane Diffusion
    :star:code
  • Bi-SSC: Geometric-Semantic Bidirectional Fusion for Camera-based 3D Semantic Scene Completion
• 场景补全
  • Unleashing Network Potentials for Semantic Scene Completion
  • Not All Voxels Are Equal: Hardness-Aware Semantic Scene Completion with Self-Distillation
    :star:code
  • Symphonize 3D Semantic Scene Completion with Contextual Instance Queries
    :star:code3D 语义
  • PaSCo: Urban 3D Panoptic Scene Completion with Uncertainty Awareness
  • Scaling Diffusion Models to Real-World 3D LiDAR Scene Completion
• 图像-图像翻译
  • StegoGAN: Leveraging Steganography for Non-Bijective Image-to-Image Translation
• 图像检测
  • LaRE^2: Latent Reconstruction Error Based Method for Diffusion-Generated Image Detection
• 图像编辑
  • Focus on Your Instruction: Fine-grained and Multi-instruction Image Editing by Attention Modulation
    :star:code
  • Emu Edit: Precise Image Editing via Recognition and Generation Tasks
  • An Edit Friendly DDPM Noise Space: Inversion and Manipulations
  • Distraction is All You Need: Memory-Efficient Image Immunization against Diffusion-Based Image Editing
  • DiffEditor: Boosting Accuracy and Flexibility on Diffusion-based Image Editing
    :star:code
  • DiffMorpher: Unleashing the Capability of Diffusion Models for Image Morphing
    :star:code
    :house:project
  • UniHuman: A Unified Model For Editing Human Images in the Wild
    :star:code
  • Contrastive Denoising Score for Text-guided Latent Diffusion Image Editing
    :house:project
  • Inversion-Free Image Editing with Language-Guided Diffusion Models
    :star:code
    :house:project
  • TiNO-Edit: Timestep and Noise Optimization for Robust Diffusion-Based Image Editing
    :star:code
  • Edit One for All: Interactive Batch Image Editing
    :star:code
    :house:project
  • SCEdit: Efficient and Controllable Image Diffusion Generation via Skip Connection Editing
    :house:project
  • On Exact Inversion of DPM-Solvers
    :star:code
    :house:project
  • Doubly Abductive Counterfactual Inference for Text-based Image Editing
    :star:code基于文本的图像编辑
  • Towards Understanding Cross and Self-Attention in Stable Diffusion for Text-Guided Image Editing
  • ZONE: Zero-Shot Instruction-Guided Local Editing
  • HIVE: Harnessing Human Feedback for Instructional Visual Editing
  • FreeDrag: Feature Dragging for Reliable Point-based Image Editing
  • The Devil is in the Details: StyleFeatureEditor for Detail-Rich StyleGAN Inversion and High Quality Image Editing
  • DragDiffusion: Harnessing Diffusion Models for Interactive Point-based Image Editing
    :star:code
  • Text-Driven Image Editing via Learnable Regions
    :star:code
    :house:project
  • LEDITS++: Limitless Image Editing using Text-to-Image Models
    :star:code
    :house:project
  • SmartEdit: Exploring Complex Instruction-based Image Editing with Multimodal Large Language Models
    :star:code
    :house:project
  • Person in Place: Generating Associative Skeleton-Guidance Maps for Human-Object Interaction Image Editing
    :star:code
    :house:project
  • PAIR-Diffusion: A Comprehensive Multimodal Object-Level Image Editor
    :star:code
    :house:project
  • Referring Image Editing: Object-level Image Editing via Referring Expressions
  • Prompt Augmentation for Self-supervised Text-guided Image Manipulation
  • Named Entity Driven Zero-Shot Image Manipulation
• 布局生成
  • Constrained Layout Generation with Factor Graphs
  • SmartMask: Context Aware High-Fidelity Mask Generation for Fine-grained Object Insertion and Layout Control
  • MaskPLAN: Masked Generative Layout Planning from Partial Input
    :star:code
  • Retrieval-Augmented Layout Transformer for Content-Aware Layout Generation
    :star:code
    :house:project
  • Visual Layout Composer: Image-Vector Dual Diffusion Model for Design Layout Generation
    :house:project
• 手写数学表达式
  • Generating Handwritten Mathematical Expressions From Symbol Graphs: An End-to-End Pipeline
• NeRF-to-NeRF
  • GenN2N: Generative NeRF2NeRF TranslationNeRF-to-NeRF
• 生成伪装图像
  • LAKE-RED: Camouflaged Images Generation by Latent Background Knowledge Retrieval-Augmented Diffusion
• 场景生成
  • Towards Realistic Scene Generation with LiDAR Diffusion Models
    :star:code
    :house:project
    :thumbsup:LiDM：首个可以根据多模态条件生成逼真的激光雷达场景方法，加速107倍
  • DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis室内场景合成
• 交互式编辑
  • Drag Your Noise: Interactive Point-based Editing via Diffusion Semantic Propagation
    :star:code
• 视频编辑
  • CCEdit: Creative and Controllable Video Editing via Diffusion Models
    :house:project
    :tv:video
  • MaskINT: Video Editing via Interpolative Non-autoregressive Masked Transformers
  • RAVE: Randomized Noise Shuffling for Fast and Consistent Video Editing with Diffusion Models
    :star:code
    :house:project
  • A Video is Worth 256 Bases: Spatial-Temporal Expectation-Maximization Inversion for Zero-Shot Video Editing
    :house:project
  • Video-P2P: Video Editing with Cross-attention Control
    :house:project
  • VidToMe: Video Token Merging for Zero-Shot Video Editing
    :house:project
  • Video Interpolation with Diffusion Models
    :star:code
  • MotionEditor: Editing Video Motion via Content-Aware Diffusion
  • CAMEL: CAusal Motion Enhancement Tailored for Lifting Text-driven Video Editing
    :star:code
  • DynVideo-E: Harnessing Dynamic NeRF for Large-Scale Motion- and View-Change Human-Centric Video Editing
    :house:project
• 漫画生成
  • The Manga Whisperer: Automatically Generating Transcriptions for Comics
    :star:code
• 文本驱动 3D 风格化
  • TeMO: Towards Text-Driven 3D Stylization for Multi-Object Meshes
• Image Warping
  • Towards Progressive Multi-Frequency Representation for Image Warping
• 图像重建
  • Equivariant Plug-and-Play Image Reconstruction
  • SPECAT: SPatial-spEctral Cumulative-Attention Transformer for High-Resolution Hyperspectral Image Reconstruction
  • Boosting Spike Camera Image Reconstruction from a Perspective of Dealing with Spike Fluctuations
• 图像拼接
  • DiffAssemble: A Unified Graph-Diffusion Model for 2D and 3D Reassembly
    :star:code
  • RecDiffusion: Rectangling for Image Stitching with Diffusion Models
    :star:code
• 姿势引导的人体图像合成
  • Coarse-to-Fine Latent Diffusion for Pose-Guided Person Image Synthesis
    :star:code
• 文本引导的人体图像合成
  • Towards Effective Usage of Human-Centric Priors in Diffusion Models for Text-based Human Image Generation
    :star:code
• 文本图像对齐
  • Text-image Alignment for Diffusion-based Perception
    :star:code
    :house:project
• 基于文本的图像色调调整
  • CLIPtone: Unsupervised Learning for Text-based Image Tone Adjustment
• 图像矢量化
  • Towards High-fidelity Artistic Image Vectorization via Texture-Encapsulated Shape Parameterization
• 文本-矢量
  • NIVeL: Neural Implicit Vector Layers for Text-to-Vector Generation
• 矢量字体
  • VecFusion: Vector Font Generation with Diffusion
• 矢量图形合成
  • SuperSVG: Superpixel-based Scalable Vector Graphics Synthesis
  • Vector Graphics Generation via Mutually Impulsed Dual-domain Diffusion
• 二维码生成
  • Text2QR: Harmonizing Aesthetic Customization and Scanning Robustness for Text-Guided QR Code Generation
    :star:code
• 背景替换
  • Relightful Harmonization: Lighting-aware Portrait Background Replacement

7.Image Progress(图像处理)

• 去鬼影
  • Generating Content for HDR Deghosting from Frequency View
• 去阴影
  • HomoFormer: Homogenized Transformer for Image Shadow Removal
• 去模糊
  • Unsupervised Blind Image Deblurring Based on Self-Enhancement
  • Latency Correction for Event-guided Deblurring and Frame Interpolation
  • LDP: Language-driven Dual-Pixel Image Defocus Deblurring Network
  • ID-Blau: Image Deblurring by Implicit Diffusion-based reBLurring AUgmentation
  • Blur2Blur: Blur Conversion for Unsupervised Image Deblurring on Unknown Domains
    :star:code
  • AdaRevD: Adaptive Patch Exiting Reversible Decoder Pushes the Limit of Image Deblurring
    :star:code
    :star:code
  • A Unified Framework for Microscopy Defocus Deblur with Multi-Pyramid Transformer and Contrastive Learning
    :star:code
• 去雾
  • ODCR: Orthogonal Decoupling Contrastive Regularization for Unpaired Image Dehazing
  • Depth Information Assisted Collaborative Mutual Promotion Network for Single Image Dehazing
  • A Semi-supervised Nighttime Dehazing Baseline with Spatial-Frequency Aware and Realistic Brightness Constraint
    :star:code
• 去噪
  • Real-World Mobile Image Denoising Dataset with Efficient Baselines
  • GenesisTex: Adapting Image Denoising Diffusion to Texture Space
  • Robust Image Denoising through Adversarial Frequency Mixup
  • Exploring Efficient Asymmetric Blind-Spots for Self-Supervised Denoising in Real-World Scenarios
  • Masked and Shuffled Blind Spot Denoising for Real-World Images
  • LAN: Learning to Adapt Noise for Image Denoising
  • Unmixing Diffusion for Self-Supervised Hyperspectral Image Denoising
  • Stable Neighbor Denoising for Source-free Domain Adaptive Segmentation
  • Transfer CLIP for Generalizable Image Denoising
  • Residual Denoising Diffusion Models
    :star:code
  • Equivariant plug-and-play image reconstruction
    :star:code
  • Patch2Self2: Self-supervised Denoising on Coresets via Matrix Sketching
  • Hyper-MD: Mesh Denoising with Customized Parameters Aware of Noise Intensity and Geometric Characteristics
  • Zero-Shot Illumination-Guided Joint Denoising and Adaptive Enhancement for Low-Light Images
    :thumbsup:中文简介
• 去雨
  • Bidirectional Multi-Scale Implicit Neural Representations for Image Deraining
    :star:code
• 去反射
  • Revisiting Single Image Reflection Removal In the Wild
  • Language-guided Image Reflection Separation图像反射分离
• 修图
  • Close Imitation of Expert Retouching for Black-and-White Photography
• 图像增强
  • Color Shift Estimation-and-Correction for Image Enhancement
  • FlowIE：Efficient Image Enhancement via Rectified Flow
  • Fourier Priors-Guided Diffusion for Zero-Shot Joint Low-Light Enhancement and Deblurring
  • Specularity Factorization for Low-Light Enhancement
  • Zero-Reference Low-Light Enhancement via Physical Quadruple Priors
    :star:code
  • Towards Robust Event-guided Low-Light Image Enhancement: A Large-Scale Real-World Event-Image Dataset and Novel Approach
    :star:code
  • Empowering Resampling Operation for Ultra-High-Definition Image Enhancement with Model-Aware Guidance
  • Light the Night: A Multi-Condition Diffusion Framework for Unpaired Low-Light Enhancement in Autonomous Driving
• 图像恢复
  • Learning Diffusion Texture Priors for Image Restoration
  • CoDe: An Explicit Content Decoupling Framework for Image Restoration
  • Wavelet-based Fourier Information Interaction with Frequency Diffusion Adjustment for Underwater Image Restoration
  • Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild
  • Look-Up Table Compression for Efficient Image Restoration
  • HIR-Diff: Unsupervised Hyperspectral Image Restoration Via Improved Diffusion Models
  • DocRes: A Generalist Model Toward Unifying Document Image Restoration Tasks
  • Image Restoration by Denoising Diffusion Models with Iteratively Preconditioned Guidance
    :star:code
  • Deep Equilibrium Diffusion Restoration with Parallel Sampling
    :star:code
  • Distilling Semantic Priors from SAM to Efficient Image Restoration Models
  • Boosting Image Restoration via Priors from Pre-trained Models
  • Adapt or Perish: Adaptive Sparse Transformer with Attentive Feature Refinement for Image Restoration
  • Selective Hourglass Mapping for Universal Image Restoration Based on Diffusion Model
    :star:code
  • Restoration by Generation with Constrained Priors
    :house:project
  • Multimodal Prompt Perceiver: Empower Adaptiveness Generalizability and Fidelity for All-in-One Image Restoration
  • Improving Image Restoration through Removing Degradations in Textual Representations
    :star:code
• 图像修复
  • Brush2Prompt: Contextual Prompt Generator for Object Inpainting
  • Don't Look into the Dark: Latent Codes for Pluralistic Image Inpainting
  • NeRFiller: Completing Scenes via Generative 3D Inpainting
  • MVIP-NeRF: Multi-view 3D Inpainting on NeRF Scenes via Diffusion Prior3D 修复
  • Structure Matters: Tackling the Semantic Discrepancy in Diffusion Models for Image Inpainting
    :star:code
• 图像超级补全
  • Shadow-Enlightened Image Outpainting
• 图像质量
  • Blind Image Quality Assessment Based on Geometric Order Learning
  • Defense Against Adversarial Attacks on No-Reference Image Quality Models with Gradient Norm Regularization
  • Bridging the Synthetic-to-Authentic Gap: Distortion-Guided Unsupervised Domain Adaptation for Blind Image Quality Assessment
  • TextCraftor: Your Text Encoder Can be Image Quality Controller
  • Boosting Image Quality Assessment through Efficient Transformer Adaptation with Local Feature Enhancement
• 恶劣天气消除
  • Genuine Knowledge from Practice: Diffusion Test-Time Adaptation for Video Adverse Weather Removal
    :star:code
  • Language-driven All-in-one Adverse Weather Removal恶劣天气消除
• 大气湍流去除
  • NB-GTR: Narrow-Band Guided Turbulence Removal
• Image Portrait Relighting(图像重照光)
  • SwitchLight: Co-design of Physics-driven Architecture and Pre-training Framework for Human Portrait Relighting
    :house:project
• 图片缩小
  • Deep Generative Model based Rate-Distortion for Image Downscaling Assessment
• 图像校正
  • Rolling Shutter Correction with Intermediate Distortion Flow Estimation
• 图像着色
  • Learning Inclusion Matching for Animation Paint Bucket Colorization
    :star:code着色
  • Automatic Controllable Colorization via Imagination
    :star:code
• 运动(去)模糊
  • Motion Blur Decomposition with Cross-shutter Guidance
  • Spike-guided Motion Deblurring with Unknown Modal Spatiotemporal Alignment
    :star:code
  • Real-World Efficient Blind Motion Deblurring via Blur Pixel Discretization
  • Efficient Multi-scale Network with Learnable Discrete Wavelet Transform for Blind Motion Deblurring
  • Motion-adaptive Separable Collaborative Filters for Blind Motion Deblurring
    :star:code
• 视频修复
  • AVID: Any-Length Video Inpainting with Diffusion Model
    :star:code
    :house:project
  • Towards Language-Driven Video Inpainting via Multimodal Large Language Models
    :house:project
• 视频去雾
  • Driving-Video Dehazing with Non-Aligned Regularization for Safety Assistance
• 视频去渲染
  • Leveraging Frame Affinity for sRGB-to-RAW Video De-rendering
• 视频去模糊
  • Frequency-aware Event-based Video Deblurring for Real-World Motion Blur
  • Blur-aware Spatio-temporal Sparse Transformer for Video Deblurring
    :star:code
    :house:project
  • FMA-Net: Flow Guided Dynamic Filtering and Iterative Feature Refinement with Multi-Attention for Joint Video Super-Resolution and Deblurring
    :star:code
    :house:project
  • DyBluRF: Dynamic Neural Radiance Fields from Blurry Monocular Video
    :house:project
• 视频增强
  • Binarized Low-light Raw Video Enhancement
• 视频质量评估
  • PTM-VQA: Efficient Video Quality Assessment Leveraging Diverse PreTrained Models from the Wild
  • Learned Scanpaths Aid Blind Panoramic Video Quality Assessment
  • Modular Blind Video Quality Assessment
  • KVQ: Kwai Video Quality Assessment for Short-form Videos
  • CPGA: Coding Priors-Guided Aggregation Network for Compressed Video Quality Enhancement
    :star:code
• 夜间颜色恒定
  • NightCC: Nighttime Color Constancy via Adaptive Channel Masking
• 照明估计
  • Towards a Perceptual Evaluation Framework for Lighting Estimation

6.Image/Video Captioning(图像/视频字幕)

• Visual Fact Checker: Enabling High-Fidelity Detailed Caption Generation
• Polos: Multimodal Metric Learning from Human Feedback for Image Captioning
  :star:code
  :house:project
• Panda-70M: Captioning 70M Videos with Multiple Cross-Modality Teachers
  :star:code
• MeaCap: Memory-Augmented Zero-shot Image Captioning
  :star:code
• Sieve: Multimodal Dataset Pruning using Image Captioning Models
• [EVCap: Retrieval-Augmented Image Captioning with External Visual--Name Memory for Open-World Comprehension]
• EVCap: Retrieval-Augmented Image Captioning with External Visual-Name Memory for Open-World Comprehension
• 视频描述/字幕
  • Streaming Dense Video Captioning
    :star:code
    :star:code
  • Video ReCap: Recursive Captioning of Hour-Long Videos
    :star:code
    :house:project
    :sunflower:dataset
  • Do You Remember? Dense Video Captioning with Cross-Modal Memory Retrieval
  • VideoCon: Robust Video-Language Alignment via Contrast Captions
    :star:code
    :house:project
  • Retrieval-Augmented Egocentric Video Captioning
• 密集字幕
  • A Picture is Worth More Than 77 Text Tokens: Evaluating CLIP-Style Models on Dense Captions
  • DIBS: Enhancing Dense Video Captioning with Unlabeled Videos via Pseudo Boundary Enrichment and Online Refinement
• 生成图解说明
  • Generating Illustrated Instructions
    :star:code
    :house:project

5.Image/Video Compression(图像/视频压缩)

• 视频压缩
  • Neural Video Compression with Feature Modulation
    :star:code
  • C3: High-Performance and Low-Complexity Neural Compression from a Single Image or Video
    :star:code
    :house:project
  • Task-Aware Encoder Control for Deep Video Compression
• 图像压缩
  • Towards Backward-Compatible Continual Learning of Image Compression
    :star:code
  • Generative Latent Coding for Ultra-Low Bitrate Image Compression
  • Dual Prior Unfolding for Snapshot Compressive Imaging
  • Enhancing Quality of Compressed Images by Mitigating Enhancement Bias Towards Compression Domain
  • SCINeRF: Neural Radiance Fields from a Snapshot Compressive Image
    :star:code
  • JDEC: JPEG Decoding via Enhanced Continuous Cosine CoefficientsJPEG 解码
  • Learned Lossless Image Compression based on Bit Plane Slicing

4.Image/Video Super-Resolution(图像超分辨率)

• Image Processing GNN: Breaking Rigidity in Super-Resolution
• Learning Large-Factor EM Image Super-Resolution with Generative Priors
• Super-Resolution Reconstruction from Bayer-Pattern Spike Streams
• Continuous Optical Zooming: A Benchmark for Arbitrary-Scale Image Super-Resolution in Real World
• Transcending the Limit of Local Window: Advanced Super-Resolution Transformer with Adaptive Token Dictionary
  :star:code
• Learning Coupled Dictionaries from Unpaired Data for Image Super-Resolution
• SinSR: Diffusion-Based Image Super-Resolution in a Single Step
  :star:code
• CAMixerSR: Only Details Need More "Attention"
• Text-guided Explorable Image Super-resolution
• CFAT: Unleashing Triangular Windows for Image Super-resolution
• SeD: Semantic-Aware Discriminator for Image Super-Resolution
• Training Generative Image Super-Resolution Models by Wavelet-Domain Losses Enables Better Control of Artifacts
• Boosting Flow-based Generative Super-Resolution Models via Learned Prior
  :star:code
• Beyond Image Super-Resolution for Image Recognition with Task-Driven Perceptual Loss
  :star:code
• AdaBM: On-the-Fly Adaptive Bit Mapping for Image Super-Resolution
  :star:code
• Uncertainty-Aware Source-Free Adaptive Image Super-Resolution with Wavelet Augmentation Transformer
• DiSR-NeRF: Diffusion-Guided View-Consistent Super-Resolution NeRF超分辨率
• Neural Super-Resolution for Real-time Rendering with Radiance Demodulation
• Self-Adaptive Reality-Guided Diffusion for Artifact-Free Super-Resolution
  :star:code
• Low-Res Leads the Way: Improving Generalization for Super-Resolution by Self-Supervised Learning
• CoSeR: Bridging Image and Language for Cognitive Super-Resolution
  :star:code
  :house:project
• Navigating Beyond Dropout: An Intriguing Solution towards Generalizable Image Super Resolution
• Bilateral Event Mining and Complementary for Event Stream Super-Resolution
• 盲图像超分辨率
  • CDFormer: When Degradation Prediction Embraces Diffusion Model for Blind Image Super-Resolution
  • A Dynamic Kernel Prior Model for Unsupervised Blind Image Super-Resolution
• 真实世界超分辨率 Universal Robustness via Median Randomized Smoothing for Real-World Super-Resolution
  • APISR: Anime Production Inspired Real-World Anime Super-Resolution
  • SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution
• VSR
  • Learning Spatial Adaptation and Temporal Coherence in Diffusion Models for Video Super-Resolution
  • Enhancing Video Super-Resolution via Implicit Resampling-based Alignment
  • Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution
    :house:project
  • Video Super-Resolution Transformer with Masked Inter&Intra-Frame Attention
    :star:code
• 文本图像超分
  • Diffusion-based Blind Text Image Super-Resolution

3.Image Classification(图像分类)

• Fair-VPT: Fair Visual Prompt Tuning for Image Classification
• Logarithmic Lenses: Exploring Log RGB Data for Image Classification
• SLICE: Stabilized LIME for Consistent Explanations for Image Classification
• Classes Are Not Equal: An Empirical Study on Image Recognition Fairness
• MCPNet: An Interpretable Classifier via Multi-Level Concept Prototypes
• SURE: SUrvey REcipes for building reliable and robust deep networks
  :star:code
• A Bayesian Approach to OOD Robustness in Image Classification
• Fourier-basis Functions to Bridge Augmentation Gap: Rethinking Frequency Augmentation in Image Classification
• Hyperspherical Classification with Dynamic Label-to-Prototype Assignment
  :star:code
• Discover and Mitigate Multiple Biased Subgroups in Image Classifiers
  :star:code
• Deep Imbalanced Regression via Hierarchical Classification Adjustment
• Large Language Models are Good Prompt Learners for Low-Shot Image Classification
  :star:code
• Modeling Collaborator: Enabling Subjective Vision Classification With Minimal Human Effort via LLM Tool-Use
• Bayesian Exploration of Pre-trained Models for Low-shot Image Classification
• Enhance Image Classification via Inter-Class Image Mixup with Diffusion Model
• Leveraging Cross-Modal Neighbor Representation for Improved CLIP Classification
  :star:code
• In-distribution Public Data Synthesis with Diffusion Models for Differentially Private Image Classification
• 域泛化图像分类
  • Leveraging Vision-Language Models for Improving Domain Generalization in Image Classification
    :house:project
• 长尾识别
  • LTGC: Long-tail Recognition via Leveraging LLMs-driven Generated Content
• 小样本图像分类
  • Frozen Feature Augmentation for Few-Shot Image Classification
• 零样本分类
  • Label Propagation for Zero-shot Classification with Vision-Language Models
    :star:code
  • CARZero: Cross-Attention Alignment for Radiology Zero-Shot Classification
    :star:code
  • Improved Zero-Shot Classification by Adapting VLMs with Text Descriptions
    :star:code零样本分类
• 细粒度
  • Fine-grained Bipartite Concept Factorization for Clustering
  • Novel Class Discovery for Ultra-Fine-Grained Visual Categorization
    :star:code
• 开集分类
  • ProTeCt: Prompt Tuning for Taxonomic Open Set Classification
• 小样本识别
  • Instance-based Max-margin for Practical Few-shot Recognition
• GCD(广义类别发现)
  • Federated Generalized Category Discovery
  • Active Generalized Category Discovery
    :star:code
  • Contrastive Mean-Shift Learning for Generalized Category Discovery
  • Solving the Catastrophic Forgetting Problem in Generalized Category Discovery

2.Image Segmentation(图像分割)

• Matching Anything by Segmenting Anything
  :star:code
• Unsupervised Universal Image Segmentation
• MESA: Matching Everything by Segmenting Anything
• MRFS: Mutually Reinforcing Image Fusion and Segmentation
• RobustSAM: Segment Anything Robustly on Degraded Images
• Hierarchical Histogram Threshold Segmentation - Auto-terminating High-detail Oversegmentation
• Multi-Space Alignments Towards Universal LiDAR Segmentation
• CoralSCOP: Segment any COral Image on this Planet分割
• SANeRF-HQ: Segment Anything for NeRF in High Quality
  :house:project
• ASAM: Boosting Segment Anything Model with Adversarial Tuning
• ODIN: A Single Model for 2D and 3D Segmentation
  :star:code
• FocSAM: Delving Deeply into Focused Objects in Segmenting Anything
  :thumbsup:摘要
• EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything
• Universal Segmentation at Arbitrary Granularity with Language Instruction通用分割
• Segment and Caption Anything
  :house:project
• COCONut: Modernizing COCO Segmentation
  :star:code
• Multi-view Aggregation Network for Dichotomous Image Segmentation
  :star:code
• OMG-Seg: Is One Model Good Enough For All Segmentation?
  :house:project
• Unsegment Anything by Simulating Deformation
• BA-SAM: Scalable Bias-Mode Attention Mask for Segment Anything Model
  :star:code
• VRP-SAM: SAM with Visual Reference Prompt
• PEM: Prototype-based Efficient MaskFormer for Image Segmentation
• Fantastic Animals and Where to Find Them: Segment Any Marine Animal with Dual SAM
  :star:code
• CLIP as RNN: Segment Countless Visual Concepts without Training Endeavor
  :house:project
• Benchmarking Segmentation Models with Mask-Preserved Attribute Editing
  :star:code
• CuVLER: Enhanced Unsupervised Object Discoveries through Exhaustive Self-Supervised Transformers
• Continual Segmentation with Disentangled Objectness Learning and Class Recognition
  :star:code
• Kandinsky Conformal Prediction: Efficient Calibration of Image Segmentation Algorithms
  :star:code
• Training-Free Open-Vocabulary Segmentation with Offline Diffusion-Augmented Prototype Generation
  :star:code
  :house:project
• Parameter Efficient Fine-tuning via Cross Block Orchestration for Segment Anything Model
• A Simple Recipe for Language-guided Domain Generalized Segmentation
  :house:project
• Rethinking Interactive Image Segmentation with Low Latency High Quality and Diverse Prompts
  :star:code
• Improving the Generalization of Segmentation Foundation Model under Distribution Shift via Weakly Supervised Adaptation
  :star:code
  :thumbsup:分割一切模型SAM泛化能力差？域适应策略给解决了
• 开放词汇分割
  • Transferable and Principled Efficiency for Open-Vocabulary Segmentation
  • USE: Universal Segment Embeddings for Open-Vocabulary Image Segmentation
  • Open-Vocabulary Segmentation with Semantic-Assisted Calibration
    :star:code
  • OVFoodSeg: Elevating Open-Vocabulary Food Image Segmentation via Image-Informed Textual Representation
• 视频分割
  • UniVS: Unified and Universal Video Segmentation with Prompts as Queries
    :star:code
  • Turb-Seg-Res: A Segment-then-Restore Pipeline for Dynamic Videos with Atmospheric Turbulence
    :house:project视频分割
  • Learning to Segment Referred Objects from Narrated Egocentric Videos
  • Decoupling Static and Hierarchical Motion Perception for Referring Video Segmentation
    :star:code
• 语义分割
  • Open Set Domain Adaptation for Semantic Segmentation
  • ContextSeg: Sketch Semantic Segmentation by Querying the Context with Attention
  • MRFP: Learning Generalizable Semantic Segmentation from Sim-2-Real with Multi-Resolution Feature Perturbation
  • TASeg: Temporal Aggregation Network for LiDAR Semantic Segmentation
  • ALGM: Adaptive Local-then-Global Token Merging for Efficient Semantic Segmentation with Plain Vision Transformers
  • HPL-ESS: Hybrid Pseudo-Labeling for Unsupervised Event-based Semantic Segmentation
  • Contextrast: Contextual Contrastive Learning for Semantic Segmentation
  • Open-Set Domain Adaptation for Semantic Segmentation
  • SG-BEV: Satellite-Guided BEV Fusion for Cross-View Semantic Segmentation
    :star:code
  • Frequency-Adaptive Dilated Convolution for Semantic Segmentation
    :star:code
  • GoodSAM: Bridging Domain and Capacity Gaps via Segment Anything Model for Distortion-aware Panoramic Semantic Segmentation
  • Improving Bird's Eye View Semantic Segmentation by Task Decomposition
    :star:code
  • UniMix: Towards Domain Adaptive and Generalizable LiDAR Semantic Segmentation in Adverse Weather
  • Open-Vocabulary Attention Maps with Token Optimization for Semantic Segmentation in Diffusion Models
  • Flattening the Parent Bias: Hierarchical Semantic Segmentation in the Poincaré Ball
  • 3D 语义分割
    • Hierarchical Intra-modal Correlation Learning for Label-free 3D Semantic Segmentation
    • OA-CNNs: Omni-Adaptive Sparse CNNs for 3D Semantic Segmentation
  • 点云语义分割
    • Rethinking Few-shot 3D Point Cloud Semantic Segmentation
      :star:code
    • PDF: A Probability-Driven Framework for Open World 3D Point Cloud Semantic Segmentation3D 点云语义分割
  • 无监督语义分割
    • Learn to Rectify the Bias of CLIP for Unsupervised Semantic Segmentation
    • EAGLE: Eigen Aggregation Learning for Object-Centric Unsupervised Semantic Segmentation
      :star:code
      :house:project
  • 小样本语义分割
    • APSeg: Auto-Prompt Network for Cross-Domain Few-Shot Semantic Segmentation
    • Unlocking the Potential of Pre-trained Vision Transformers for Few-Shot Semantic Segmentation through Relationship Descriptors
  • 零样本语义分割
    • Exploring Regional Clues in CLIP for Zero-Shot Semantic Segmentation
  • 半监督语义分割
    • Training Vision Transformers for Semi-Supervised Semantic Segmentation
    • Density-Guided Semi-Supervised 3D Semantic Segmentation with Dual-Space Hardness Sampling
    • AllSpark: Reborn Labeled Features from Unlabeled in Transformer for Semi-Supervised Semantic Segmentation
      :star:code
    • CorrMatch: Label Propagation via Correlation Matching for Semi-Supervised Semantic Segmentation
      :star:code
    • Towards the Uncharted: Density-Descending Feature Perturbation for Semi-supervised Semantic Segmentation
      :star:code
    • RankMatch: Exploring the Better Consistency Regularization for Semi-supervised Semantic Segmentation
  • 弱监督语义分割
    • Class Tokens Infusion for Weakly Supervised Semantic Segmentation
    • Frozen CLIP: A Strong Backbone for Weakly Supervised Semantic Segmentation
    • DuPL: Dual Student with Trustworthy Progressive Learning for Robust Weakly Supervised Semantic Segmentation
      :star:code
    • Hunting Attributes: Context Prototype-Aware Learning for Weakly Supervised Semantic Segmentation
      :star:code
    • Separate and Conquer: Decoupling Co-occurrence via Decomposition and Representation for Weakly Supervised Semantic Segmentation
      :star:code
    • PSDPM: Prototype-based Secondary Discriminative Pixels Mining for Weakly Supervised Semantic Segmentation
    • From SAM to CAMs: Exploring Segment Anything Model for Weakly Supervised Semantic Segmentation
  • 域泛化语义分割
    • Collaborating Foundation Models for Domain Generalized Semantic Segmentation
      :star:code
    • Style Blind Domain Generalized Semantic Segmentation via Covariance Alignment and Semantic Consistence Contrastive Learning
      :star:code
    • Stronger Fewer & Superior: Harnessing Vision Foundation Models for Domain Generalized Semantic Segmentation
      :star:code
  • 文本监督语义分割
    • Image-Text Co-Decomposition for Text-Supervised Semantic Segmentation
  • 开放世界语义分割
    • Open-World Semantic Segmentation Including Class Similarity
      :star:code
  • 开放词汇语义分割
    • Open-Vocabulary 3D Semantic Segmentation with Foundation Models
    • Open-Vocabulary Semantic Segmentation with Image Embedding Balancing
    • CAT-Seg: Cost Aggregation for Open-Vocabulary Semantic Segmentation
    • Image-to-Image Matching via Foundation Models: A New Perspective for Open-Vocabulary Semantic Segmentation
    • SED: A Simple Encoder-Decoder for Open-Vocabulary Semantic Segmentation
      :star:code
    • Emergent Open-Vocabulary Semantic Segmentation from Off-the-shelf Vision-Language Models
      :star:code
• 全景分割
  • Semantics Distortion and Style Matter: Towards Source-free UDA for Panoramic Segmentation
  • ECLIPSE: Efficient Continual Learning in Panoptic Segmentation with Visual Prompt Tuning
    :star:code
  • PanoOcc: Unified Occupancy Representation for Camera-based 3D Panoptic Segmentation
    :star:code
  • Task-aligned Part-aware Panoptic Segmentation through Joint Object-Part Representations
    :star:code
• 实例分割
  • Extreme Point Supervised Instance Segmentation
  • Mudslide: A Universal Nuclear Instance Segmentation Method
  • Semantic-aware SAM for Point-Prompted Instance Segmentation
  • SAI3D: Segment Any Instance in 3D Scenes
    :house:project
  • DiverGen: Improving Instance Segmentation by Learning Wider Data Distribution with More Diverse Generative Data
  • FISBe: A Real-World Benchmark Dataset for Instance Segmentation of Long-Range Thin Filamentous Structures
    :star:code
  • Teeth-SEG: An Efficient Instance Segmentation Framework for Orthodontic Treatment based on Multi-Scale Aggregation and Anthropic Prior Knowledge
  • 开放词汇实例分割
    • MaskClustering: View Consensus based Mask Graph Clustering for Open-Vocabulary 3D Instance Segmentation
      :house:project
  • 3D 实例分割
    • BSNet: Box-Supervised Simulation-assisted Mean Teacher for 3D Instance Segmentation
      :star:code
    • Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance
    • Edge-Aware 3D Instance Segmentation Network with Intelligent Semantic Prior
    • UnScene3D: Unsupervised 3D Instance Segmentation for Indoor Scenes
• 场景分割
  • No Time to Train: Empowering Non-Parametric Networks for Few-shot 3D Scene Segmentation
    :star:code
  • MirageRoom: 3D Scene Segmentation with 2D Pre-trained Models by Mirage Projection
• 动作分割
  • Progress-Aware Online Action Segmentation for Egocentric Procedural Task Videos
  • Coherent Temporal Synthesis for Incremental Action Segmentation
  • Efficient and Effective Weakly-Supervised Action Segmentation via Action-Transition-Aware Boundary Alignment
  • Temporally Consistent Unbalanced Optimal Transport for Unsupervised Action Segmentation
  • FACT: Frame-Action Cross-Attention Temporal Modeling for Efficient Action Segmentation
    :star:code
• 参考图像分割
  • LQMFormer: Language-aware Query Mask Transformer for Referring Image Segmentation
  • Mask Grounding for Referring Image Segmentation
    :house:project
  • Curriculum Point Prompting for Weakly-Supervised Referring Image Segmentation
  • Prompt-Driven Referring Image Segmentation with Instance Contrasting
• 指代表达式分割
  • Unveiling Parts Beyond Objects: Towards Finer-Granularity Referring Expression Segmentation
    :star:code
• VOS
  • Point-VOS: Pointing Up Video Object Segmentation
    :house:project
  • Dual Prototype Attention for Unsupervised Video Object Segmentation
    :star:code
  • Depth-aware Test-Time Training for Zero-shot Video Object Segmentation
    :star:code
  • Putting the Object Back into Video Object Segmentation
    :house:project
  • Event-assisted Low-Light Video Object Segmentation
  • Guided Slot Attention for Unsupervised Video Object Segmentation
  • LoSh: Long-Short Text Joint Prediction Network for Referring Video Object Segmentation
    :star:code
  • RMem: Restricted Memory Banks Improve Video Object Segmentation
• VSS
  • Infer from What You Have Seen Before: Temporally-dependent Classifier for Semi-supervised Video Semantic Segmentation
  • Vanishing-Point-Guided Video Semantic Segmentation of Driving Scenes
• VIS
  • VideoCutLER: Surprisingly Simple Unsupervised Video Instance Segmentation
• 抠图
  • In-Context Matting
    :star:code
  • Unifying Automatic and Interactive Matting with Pretrained ViTs
  • MaGGIe: Masked Guided Gradual Human Instance Matting
  • EFormer: Enhanced Transformer towards Semantic-Contour Features of Foreground for Portraits Matting
• 少样本分割
  • Rethinking Prior Information Generation with CLIP for Few-Shot Segmentation
  • Domain-Rectifying Adapter for Cross-Domain Few-Shot Segmentation
    :star:code
  • Visual Prompting for Generalized Few-shot Segmentation: A Multi-scale Approach
  • Adapt Before Comparison: A New Perspective on Cross-Domain Few-Shot Segmentation
  • LLaFS: When Large Language Models Meet Few-Shot Segmentation
  • Cross-Domain Few-Shot Segmentation via Iterative Support-Query Correspondence Mining
    :star:code
  • Addressing Background Context Bias in Few-Shot Segmentation through Iterative Modulation
• 零样本分割
  • Diffuse Attend and Segment: Unsupervised Zero-Shot Segmentation using Stable Diffusion
    :star:code
• 裂纹分割
  • Mind Marginal Non-Crack Regions: Clustering-Inspired Representation Learning for Crack Segmentation
• 交互式分割
  • GraCo: Granularity-Controllable Interactive Segmentation
    :tv:video
    :thumbsup:摘要
  • MFP: Making Full Use of Probability Maps for Interactive Image Segmentation
    :star:code
• 无模态分割
  • pix2gestalt: Amodal Segmentation by Synthesizing Wholes
    :star:code
    :house:project
• 3D 分割
  • OmniSeg3D: Omniversal 3D Segmentation via Hierarchical Contrastive Learning
  • PartDistill: 3D Shape Part Segmentation by Vision-Language Model Distillation
  • Cross-Dimension Affinity Distillation for 3D EM Neuron Segmentation
  • LASO: Language-guided Affordance Segmentation on 3D Object

1.其它(other)

• Implicit Motion Function

• Rewrite the stars

• Adapters Strike Back
  :star:code

• Detector-Free Structure from Motion

• Utility-Fairness Trade-Offs and How to Find Them

• Learning Degradation-Independent Representations for Camera ISP Pipelines

• Beyond First-Order Tweedie: Solving Inverse Problems using Latent Diffusion

• Event-based Visible and Infrared Fusion via Multi-task Collaboration

• DemoCaricature: Democratising Caricature Generation with a Rough Sketch

• PolarRec: Improving Radio Interferometric Data Reconstruction Using Polar Coordinates

• Generative Multimodal Models are In-Context Learners

• CLOVA: A Closed-LOop Visual Assistant with Tool Usage and Update

• Fully Exploiting Every Real Sample: SuperPixel Sample Gradient Model Stealing

• On the Content Bias in Frechet Video Distance

• Beyond Seen Primitive Concepts and Attribute-Object Compositional Learning

• Rotation-Agnostic Image Representation Learning for Digital Pathology

• Promptable Behaviors: Personalizing Multi-Objective Rewards from Human Preferences

• Neural Lineage

• Scaled Decoupled Distillation

• Multiview Aerial Visual RECognition (MAVREC): Can Multi-view Improve Aerial Visual Perception?

• VAREN: Very Accurate and Realistic Equine Network

• Passive Snapshot Coded Aperture Dual-Pixel RGB-D Imaging

• Generative Proxemics: A Prior for 3D Social Interaction from Images

• Non-Rigid Structure-from-Motion: Temporally-Smooth Procrustean Alignment and Spatially-Variant Deformation Modeling

• Mitigating Noisy Correspondence by Geometrical Structure Consistency Learning

• General Point Model Pretraining with Autoencoding and Autoregressive

• Estimating Extreme 3D Image Rotations using Cascaded Attention

• Fitting Flats to Flats

• Generative Powers of Ten

• Identifying Important Group of Pixels using Interactions

• ParameterNet: Parameters Are All You Need for Large-scale Visual Pretraining of Mobile Networks

• HIT: Estimating Internal Human Implicit Tissues from the Body Surface

• Taming the Tail in Class-Conditional GANs: Knowledge Sharing via Unconditional Training at Lower Resolutions

• CAPE: CAM as a Probabilistic Ensemble for Enhanced DNN Interpretation

• Fooling Polarization-Based Vision using Locally Controllable Polarizing Projection

• Total Selfie: Generating Full-Body Selfies

• Why Not Use Your Textbook? Knowledge-Enhanced Procedure Planning of Instructional Videos

• Pixel-Aligned Language Model

• ReCoRe: Regularized Contrastive Representation Learning of World Model

• Self-Calibrating Vicinal Risk Minimisation for Model Calibration

• From a Bird's Eye View to See: Joint Camera and Subject Registration without the Camera Calibration

• Motion Diversification Networks

• Ungeneralizable Examples

• Generalized Event Cameras

• Event-based Structure-from-Orbit

• Seeing the World through Your Eyes

• ProMotion: Prototypes As Motion Learners

• Move Anything with Layered Scene Diffusion

• GLACE: Global Local Accelerated Coordinate Encoding
  :star:code
  :house:project

• Quantifying Task Priority for Multi-Task Optimization

• Model Adaptation for Time Constrained Embodied Control

• Objects as Volumes: A Stochastic Geometry View of Opaque Solids

• EvDiG: Event-guided Direct and Global Components Separation

• Efficient Model Stealing Defense with Noise Transition Matrix

• OpenStreetView-5M: The Many Roads to Global Visual Geolocation

• WaveMo: Learning Wavefront Modulations to See Through Scattering

• All Rivers Run to the Sea: Private Learning with Asymmetric Flows

• HDQMF: Holographic Feature Decomposition Using Quantum Algorithms

• Cross-dimension Affinity Distillation for 3D EM Neuron Segmentation

• READ: Retrieval-Enhanced Asymmetric Diffusion for Motion Planning

• AssistGUI: Task-Oriented PC Graphical User Interface Automation

• Towards Robust Learning to Optimize with Theoretical Guarantees

• Selective Nonlinearities Removal from Digital Signals

• Ensemble Diversity Facilitates Adversarial Transferability

• One-dimensional Adapter to Rule Them All: Concepts Diffusion Models and Erasing Applications

• DiG-IN: Diffusion Guidance for Investigating Networks - Uncovering Classifier Differences Neuron Visualisations and Visual Counterfactual Explanations

• Spatial-Aware Regression for Keypoint Localization

• LP++: A Surprisingly Strong Linear Probe for Few-Shot CLIP

• One More Step: A Versatile Plug-and-Play Module for Rectifying Diffusion Schedule Flaws and Enhancing Low-Frequency Controls

• Online Task-Free Continual Generative and Discriminative Learning via Dynamic Cluster Memory

• Smart Help: Strategic Opponent Modeling for Proactive and Adaptive Robot Assistance in Households

• SPOT: Self-Training with Patch-Order Permutation for Object-Centric Learning with Autoregressive Transformers

• The Mirrored Influence Hypothesis: Efficient Data Influence Estimation by Harnessing Forward Passes

• Segment Any Event Streams via Weighted Adaptation of Pivotal Tokens

• The More You See in 2D the More You Perceive in 3D

• Versatile Navigation Under Partial Observability via Value-guided Diffusion Policy

• DiffusionLight: Light Probes for Free by Painting a Chrome Ball

• Masked Spatial Propagation Network for Sparsity-Adaptive Depth Refinement

• Probabilistic Sampling of Balanced K-Means using Adiabatic Quantum Computing

• QUADify: Extracting Meshes with Pixel-level Details and Materials from Images

• Mirasol3B: A Multimodal Autoregressive Model for Time-Aligned and Contextual Modalities

• E-GPS: Explainable Geometry Problem Solving via Top-Down Solver and Bottom-Up Generator

• Zero-Shot Structure-Preserving Diffusion Model for High Dynamic Range Tone Mapping

• Outdoor Scene Extrapolation with Hierarchical Generative Cellular Automata

• Leveraging Camera Triplets for Efficient and Accurate Structure-from-Motion

• Partial-to-Partial Shape Matching with Geometric Consistency

• Representing Part-Whole Hierarchies in Foundation Models by Learning Localizability Composability and Decomposability from Anatomy via Self Supervision

• Interpretable Measures of Conceptual Similarity by Complexity-Constrained Descriptive Auto-Encoding

• Resolution Limit of Single-Photon LiDAR

• Permutation Equivariance of Transformers and Its Applications

• From Audio to Photoreal Embodiment: Synthesizing Humans in Conversations

• Continual-MAE: Adaptive Distribution Masked Autoencoders for Continual Test-Time Adaptation

• GDA: Generalized Diffusion for Robust Test-time Adaptation

• Telling Left from Right: Identifying Geometry-Aware Semantic Correspondence

• Amodal Ground Truth and Completion in the Wild

• Spatio-Temporal Turbulence Mitigation: A Translational Perspective

• EgoGen: An Egocentric Synthetic Data Generator

• Time- Memory- and Parameter-Efficient Visual Adaptation

• Finsler-Laplace-Beltrami Operators with Application to Shape Analysis

• Intensity-Robust Autofocus for Spike Camera

• LASA: Instance Reconstruction from Real Scans using A Large-scale Aligned Shape Annotation Dataset

• Learning to Navigate Efficiently and Precisely in Real Environments

• State Space Models for Event Cameras

• Real-Time Exposure Correction via Collaborative Transformations and Adaptive Sampling

• Robust Self-calibration of Focal Lengths from the Fundamental Matrix

• Anomaly Score: Evaluating Generative Models and Individual Generated Images based on Complexity and Vulnerability

• PromptCoT: Align Prompt Distribution via Adapted Chain-of-Thought

• Uncertainty-Guided Never-Ending Learning to Drive

• Flow-Guided Online Stereo Rectification for Wide Baseline Stereo

• Don't Drop Your Samples! Coherence-Aware Training Benefits Conditional Diffusion

• Improving Spectral Snapshot Reconstruction with Spectral-Spatial Rectification

• Data-Efficient Multimodal Fusion on a Single GPU

• Learning the 3D Fauna of the Web

• iToF-flow-based High Frame Rate Depth Imaging

• Adaptive Softassign via Hadamard-Equipped Sinkhorn

• Step Differences in Instructional Video

• EASE-DETR: Easing the Competition among Object Queries

• Making Visual Sense of Oracle Bones for You and Me

• 2S-UDF: A Novel Two-stage UDF Learning Method for Robust Non-watertight Model Reconstruction from Multi-view Images多视图图像重建

• Multimodal Representation Learning by Alternating Unimodal Adaptation多模态

• Omni-SMoLA: Boosting Generalist Multimodal Models with Soft Mixture of Low-rank Experts

• Multi-Modal Proxy Learning Towards Personalized Visual Multiple Clustering多模态

• Efficient Hyperparameter Optimization with Adaptive Fidelity Identification

• Multi-modal learning for geospatial vegetation forecasting
  :star:code

• AM-RADIO: Agglomerative Vision Foundation Model Reduce All Domains Into One
  :star:code

• Practical Measurements of Translucent Materials with Inter-Pixel Translucency Prior

• Towards General Robustness Verification of MaxPool-based Convolutional Neural Networks via Tightening Linear Approximation
  :star:code

• Non-rigid Structure-from-Motion: Temporally-smooth Procrustean Alignment and Spatially-variant Deformation Modeling

• Seeing Motion at Nighttime with an Event Camera
  :star:code

• Batch Normalization Alleviates the Spectral Bias in Coordinate Networks

• Affine Equivariant Networks Based on Differential Invariants

• NC-TTT: A Noise Constrastive Approach for Test-Time Training

• Infinigen Indoors: Photorealistic Indoor Scenes using Procedural Generation

• Unexplored Faces of Robustness and Out-of-Distribution: Covariate Shifts in Environment and Sensor Domains

• Pre-training Vision Models with Mandelbulb Variations

• Noisy One-point Homographies are Surprisingly Good

• Revisiting Global Translation Estimation with Feature Tracks
  :house:project

• Efficient Scene Recovery Using Luminous Flux Prior

• MR-VNet: Media Restoration using Volterra Networks

• LEAD: Exploring Logit Space Evolution for Model Selection

• EventPS: Real-Time Photometric Stereo Using an Event Camera

• Your Transferability Barrier is Fragile: Free-Lunch for Transferring the Non-Transferable Learning

• A Theory of Joint Light and Heat Transport for Lambertian Scenes

• A Physics-informed Low-rank Deep Neural Network for Blind and Universal Lens Aberration Correction

• MCNet: Rethinking the Core Ingredients for Accurate and Efficient Homography Estimation

• Animating General Image with Large Visual Motion Model

• Pixel-level Semantic Correspondence through Layout-aware Representation Learning and Multi-scale Matching Integration

• Tuning Stable Rank Shrinkage: Aiming at the Overlooked Structural Risk in Fine-tuning

• Domain Gap Embeddings for Generative Dataset Augmentation

• Absolute Pose from One or Two Scaled and Oriented Features

• Small Steps and Level Sets: Fitting Neural Surface Models with Point Guidance

• From Variance to Veracity: Unbundling and Mitigating Gradient Variance in Differentiable Bundle Adjustment Layers

• Navigate Beyond Shortcuts: Debiased Learning Through the Lens of Neural Collapse

• Latent Modulated Function for Computational Optimal Continuous Image Representation

• Improved Self-Training for Test-Time Adaptation

• Learning with Structural Labels for Learning with Noisy Labels

• An N-Point Linear Solver for Line and Motion Estimation with Event Cameras

• Not All Classes Stand on Same Embeddings: Calibrating a Semantic Distance with Metric Tensor

• Hybrid Proposal Refiner: Revisiting DETR Series from the Faster R-CNN Perspective

• FADES: Fair Disentanglement with Sensitive Relevance

• D^4: Dataset Distillation via Disentangled Diffusion Model

• Embracing Unimodal Aleatoric Uncertainty for Robust Multimodal Fusion

• Adversarial Distillation Based on Slack Matching and Attribution Region Alignment

• In Search of a Data Transformation That Accelerates Neural Field Training

• SIRA: Scalable Inter-frame Relation and Association for Radar Perception

• Learning Discriminative Dynamics with Label Corruption for Noisy Label Detection

• Efficient Detection of Long Consistent Cycles and its Application to Distributed Synchronization

• Shadows Don't Lie and Lines Can't Bend! Generative Models don't know Projective Geometry...for now

• Explaining the Implicit Neural Canvas: Connecting Pixels to Neurons by Tracing their Contributions

• Decompose-and-Compose: A Compositional Approach to Mitigating Spurious Correlation

• Morphological Prototyping for Unsupervised Slide Representation Learning in Computational Pathology

• Tune-An-Ellipse: CLIP Has Potential to Find What You Want

• Task-Driven Wavelets using Constrained Empirical Risk Minimization

• TurboSL: Dense Accurate and Fast 3D by Neural Inverse Structured Light

• Diffeomorphic Template Registration for Atmospheric Turbulence Mitigation

• Robust Noisy Correspondence Learning with Equivariant Similarity Consistency

• Scaling Laws for Data Filtering-- Data Curation cannot be Compute Agnostic
  :star:code

• Learning to Rank Patches for Unbiased Image Redundancy Reduction

• Plug-and-Play Diffusion Distillation

• Minimal Perspective Autocalibration

• Differentiable Micro-Mesh Construction

• PointBeV: A Sparse Approach for BeV Predictions
  :star:code

• Sheared Backpropagation for Fine-tuning Foundation Models

• AirPlanes: Accurate Plane Estimation via 3D-Consistent Embeddings

• Differentiable Neural Surface Refinement for Modeling Transparent Objects

• Communication-Efficient Collaborative Perception via Information Filling with Codebook

• Friendly Sharpness-Aware Minimization
  :star:code

• RepAn: Enhanced Annealing through Re-parameterization
  :thumbsup:摘要

• LiSA：LiDAR Localization with Semantic Awareness
  :thumbsup:摘要

• Deciphering ‘What’ and ‘Where’ Visual Pathways from Spectral Clustering of Layer-Distributed Neural Representations

• As-Plausible-As-Possible: Plausibility-Aware Mesh Deformation Using 2D Diffusion Priors
  :house:project

• Fun with Flags: Robust Principal Directions via Flag Manifolds

• Steerers: A Framework for Rotation Equivariant Keypoint Descriptors
  :star:code

• GOV-NeSF: Generalizable Open-Vocabulary Neural Semantic Fields

• PixelRNN: In-pixel Recurrent Neural Networks for End-to-end-optimized Perception with Neural Sensors

• InceptionNeXt: When Inception Meets ConvNeXt
  :star:code

• Aligning and Prompting Everything All at Once for Universal Visual Perception
  :thumbsup:摘要

• Backpropagation-free Network for 3D Test-Time Adaptation
  :star:code

• Accept the Modality Gap: An Exploration in the Hyperbolic Space

• Discontinuity-preserving Normal Integration with Auxiliary Edges
  :tv:video

• 1-Lipschitz Layers Compared: Memory Speed and Certifiable Robustness
  :star:code

• PoNQ: a Neural QEM-based Mesh Representation

• Generating Non-Stationary Textures using Self-Rectification
  :star:code

• Degrees of Freedom Matter: Inferring Dynamics from Point Trajectories
  :house:project

• Multi-agent Collaborative Perception via Motion-aware Robust Communication Network

• ActiveDC: Distribution Calibration for Active Finetuning

• Spanning Training Progress: Temporal Dual-Depth Scoring (TDDS) for Enhanced Dataset Pruning

• Binding Touch to Everything: Learning Unified Multimodal Tactile Representations
  :house:project

• DiffCast: A Unified Framework via Residual Diffusion for Precipitation Nowcasting
  :star:code降水临近预报

• Intrinsic Image Diffusion for Indoor Single-view Material Estimation
  :house:project室内单视图材料估计

• Neural Underwater Scene Representation

• UniPTS: A Unified Framework for Proficient Post-Training Sparsity
  :thumbsup:摘要

• De-Diffusion Makes Text a Strong Cross-Modal Interface
  :house:project

• MMM: Generative Masked Motion Model
  :house:project

• Gaussian Shadow Casting for Neural Characters

• ExMap: Leveraging Explainability Heatmaps for Unsupervised Group Robustness to Spurious Correlations
  :star:code

• Learning Structure-from-Motion with Graph Attention Networks

• VINECS: Video-based Neural Character Skinning

• CrossMAE: Cross-Modality Masked Autoencoders for Region-Aware Audio-Visual Pre-Training

• PolarMatte: Fully Computational Ground-Truth-Quality Alpha Matte Extraction for Images and Video using Polarized Screen Matting

• Diffusion 3D Features (Diff3F): Decorating Untextured Shapes with Distilled Semantic Features

• Mind Artist: Creating Artistic Snapshots with Human Thought

• ViT-Lens: Towards Omni-modal Representations
  :star:code

• Florence-2: Advancing a Unified Representation for a Variety of Vision Tasks

• Mask4Align: Aligned Entity Prompting with Color Masks for Multi-Entity Localization Problems

• Locally Adaptive Neural 3D Morphable Models
  :star:code

• SignGraph: A Sign Sequence is Worth Graphs of Nodes

• PELA: Learning Parameter-Efficient Models with Low-Rank Approximation
  :star:code

• Versatile Navigation under Partial Observability via Value-Guided Diffusion Policy

• Dr2Net: Dynamic Reversible Dual-Residual Networks for Memory-Efficient Finetuning

• Discovering and Mitigating Visual Biases through Keyword Explanation

• PLGSLAM: Progressive Neural Scene Represenation with Local to Global Bundle Adjustment

• Generalized Large-Scale Data Condensation via Various Backbone and Statistical Matching
  :star:code数据压缩

• L2B: Learning to Bootstrap Robust Models for Combating Label Noise
  :star:code

• GART: Gaussian Articulated Template Models
  :house:project

• Towards Learning a Generalist Model for Embodied Navigation

• Revisiting Sampson Approximations for Geometric Estimation Problems

• Real-Time Neural BRDF with Spherically Distributed Primitives

• PIGEON: Predicting Image Geolocations图像地理位置

• Uncertainty Visualization via Low-Dimensional Posterior Projections

• Eclipse: Disambiguating Illumination and Materials using Unintended Shadows
  :house:project

• Fast ODE-based Sampling for Diffusion Models in Around 5 Steps
  :star:code

• CLiC: Concept Learning in Context

• Pick-or-Mix: Dynamic Channel Sampling for ConvNets

• AutoAD III: The Prequel - Back to the Pixels
  :house:project

• Training-free Pretrained Model Merging
  :star:code

• Overcoming Generic Knowledge Loss with Selective Parameter Update

• Selective nonlinearities removal from digital signals

• Memory-Scalable and Simplified Functional Map Learning

• Fully Exploiting Every Real Sample: Super-Pixel Sample Gradient Model Stealing

• Hierarchical Correlation Clustering and Tree Preserving Embedding

• GLID: Pre-training a Generalist Encoder-Decoder Vision Model

• SynSP: Synergy of Smoothness and Precision in Pose Sequences Refinement

• MS-DETR: Efficient DETR Training with Mixed Supervision

• PortraitBooth: A Versatile Portrait Model for Fast Identity-preserved Personalization
  :thumbsup:摘要

• Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data
  :star:code

• Holodeck: Language Guided Generation of 3D Embodied AI Environments

• Unified Entropy Optimization for Open-Set Test-Time Adaptation

• IIRP-Net: Iterative Inference Residual Pyramid Network for Enhanced Image Registration图像配准

• H-ViT: A Hierarchical Vision Transformer for Deformable Image Registration

• Generative Unlearning for Any Identity
  :star:code

• Error Detection in Egocentric Procedural Task Videos

• Enhancing Multimodal Cooperation via Sample-level Modality Valuation
  :star:code

• Task2Box: Box Embeddings for Modeling Asymmetric Task Relationships

• Ink Dot-Oriented Differentiable Optimization for Neural Image Halftoning

• SVDTree: Semantic Voxel Diffusion for Single Image Tree Reconstruction
  :star:code单图像树重建

• Enhancing Intrinsic Features for Debiasing via Investigating Class-Discerning Common Attributes in Bias-Contrastive Pair去偏见

• BoQ: A Place is Worth a Bag of Learnable Queries

• Distilled Datamodel with Reverse Gradient Matching

• Towards Calibrated Multi-label Deep Neural Networks

• BEHAVIOR Vision Suite: Customizable Dataset Generation via Simulation

• MART: Masked Affective RepresenTation Learning via Masked Temporal Distribution Distillation

• Gradient-based Parameter Selection for Efficient Fine-Tuning

• In2SET: Intra-Inter Similarity Exploiting Transformer for Dual-Camera Compressive Hyperspectral Imaging

• ChAda-ViT : Channel Adaptive Attention for Joint Representation Learning of Heterogeneous Microscopy Images

• Stationary Representations: Optimally Approximating Compatibility and Implications for Improved Model Replacements

• Summarize the Past to Predict the Future: Natural Language Descriptions of Context Boost Multimodal Object Interaction Anticipation
  :star:code
  :house:project

• SVDinsTN: A Tensor Network Paradigm for Efficient Structure Search from Regularized Modeling Perspective

• Masked Autoencoders for Microscopy are Scalable Learners of Cellular Biology

• Epistemic Uncertainty Quantification For Pre-trained Neural Network

• Fooling Polarization-based Vision using Locally Controllable Polarizing Projection

• TEA: Test-time Energy Adaptation
  :star:code

• Would Deep Generative Models Amplify Bias in Future Models?

• A Subspace-Constrained Tyler's Estimator and its Applications to Structure from Motion

• Domain-Specific Block Selection and Paired-View Pseudo-Labeling for Online Test-Time Adaptation
  :star:code

• DeMatch: Deep Decomposition of Motion Field for Two-View Correspondence Learning
  :star:code

• Explaining CLIP's Performance Disparities on Data from Blind/Low Vision Users

• Make Me a BNN: A Simple Strategy for Estimating Bayesian Uncertainty from Pre-trained Models

• CURSOR: Scalable Mixed-Order Hypergraph Matching with CUR Decomposition

• Bayesian Differentiable Physics for Cloth Digitalization
  :star:code

• DIMAT: Decentralized Iterative Merging-And-Training for Deep Learning Models

• FINER: Flexible Spectral-bias Tuning in Implicit NEural Representation by Variable-periodic Activation Functions

• Physical Property Understanding from Language-Embedded Feature Fields
  :star:code

• Clustering for Protein Representation Learning

• Learning Triangular Distribution in Visual World

• InstructDiffusion: A Generalist Modeling Interface for Vision Tasks

• NeISF: Neural Incident Stokes Field for Geometry and Material Estimation

• Correspondence-Free Non-Rigid Point Set Registration Using Unsupervised Clustering Analysis

• Robust Depth Enhancement via Polarization Prompt Fusion Tuning
  :star:code

• Dual-Scale Transformer for Large-Scale Single-Pixel Imaging
  :star:code

• Posterior Distillation Sampling
  :house:project

• Spin-UP: Spin Light for Natural Light Uncalibrated Photometric Stereo
  :star:code

• Efficient Deformable ConvNets: Rethinking Dynamic and Sparse Operator for Vision Applications
  :star:code

• MaxQ: Multi-Axis Query for N:M Sparsity Network
  :star:code

• AETTA: Label-Free Accuracy Estimation for Test-Time Adaptation
  :star:code

• Can Biases in ImageNet Models Explain Generalization?
  :star:code

• LLaMA-Excitor: General Instruction Tuning via Indirect Feature Interaction
  :house:project

• From Activation to Initialization: Scaling Insights for Optimizing Neural Fields

• Exploiting Inter-sample and Inter-feature Relations in Dataset Distillation
  :star:code

• UDiFF: Generating Conditional Unsigned Distance Fields with Optimal Wavelet Diffusion
  :star:code

• PikeLPN: Mitigating Overlooked Inefficiencies of Low-Precision Neural Networks

• PredToken: Predicting Unknown Tokens and Beyond with Coarse-to-Fine Iterative Decoding

• AdaShift: Learning Discriminative Self-Gated Neural Feature Activation With an Adaptive Shift Factor

• Sparse Views, Near Light: A Practical Paradigm for Uncalibrated Point-light Photometric Stereo

• MGMap: Mask-Guided Learning for Online Vectorized HD Map Construction
  :star:code

• Prompt Learning via Meta-Regularization
  :star:code

• Scalable 3D Registration via Truncated Entry-wise Absolute Residuals

• CHAIN: Enhancing Generalization in Data-Efficient GANs via lipsCHitz continuity constrAIned Normalization

• Alpha-CLIP: A CLIP Model Focusing on Wherever You Want
  :star:code
  :house:project

• Generative Quanta Color Imaging
  :star:code

• Lift3D: Zero-Shot Lifting of Any 2D Vision Model to 3D

• MedBN: Robust Test-Time Adaptation against Malicious Test Samples

• Material Palette: Extraction of Materials from a Single Image
  :star:code
  :house:project

• Adaptive Random Feature Regularization on Fine-tuning Deep Neural Networks

• Towards Large-scale 3D Representation Learning with Multi-dataset Point Prompt Training
  :star:code

• Riemannian Multinomial Logistics Regression for SPD Neural Networks
  :star:code

• A&B BNN: Add&Bit-Operation-Only Hardware-Friendly Binary Neural Network
  :star:code

• BiPer: Binary Neural Networks using a Periodic Function

• Backpropagation-free Network for 3D Test-time Adaptation
  :star:code

• Estimating Noisy Class Posterior with Part-level Labels for Noisy Label Learning
  :star:code

• ImageNet-D: Benchmarking Neural Network Robustness on Diffusion Synthetic Object
  :star:code

• Region-Based Representations Revisited

• Neural Clustering based Visual Representation Learning
  :star:code

• Efficient Stitchable Task Adaptation
  :star:code

• Pose-Guided Self-Training with Two-Stage Clustering for Unsupervised Landmark Discovery

• Laplacian-guided Entropy Model in Neural Codec with Blur-dissipated Synthesis

• PeLK: Parameter-efficient Large Kernel ConvNets with Peripheral Convolution

• Frequency Decoupling for Motion Magnification via Multi-Level Isomorphic Architecture
  :star:code

• LSK3DNet: Towards Effective and Efficient 3D Perception with Large Sparse Kernels
  :star:code

• Continual Forgetting for Pre-trained Vision Models
  :star:code

• EarthLoc: Astronaut Photography Localization by Indexing Earth from Space
  :star:code

• SpikingResformer: Bridging ResNet and Vision Transformer in Spiking Neural Networks

• Learned Trajectory Embedding for Subspace Clustering

• UnO: Unsupervised Occupancy Fields for Perception and Forecasting

• ParamISP: Learned Forward and Inverse ISPs using Camera Parameters

• PNeRV: Enhancing Spatial Consistency via Pyramidal Neural Representation for Videos

• Learning Spatial Features from Audio-Visual Correspondence in Egocentric Videos
  :house:project

• Towards More Accurate Diffusion Model Acceleration with A Timestep Aligner

• Learning Object State Changes in Videos: An Open-World Perspective
  :house:project

• U-VAP: User-specified Visual Appearance Personalization via Decoupled Self Augmentation

• Improved Implicit Neural Representation with Fourier Reparameterized Training

• Modeling Multimodal Social Interactions: New Challenges and Baselines with Densely Aligned Representations

• AlignMiF: Geometry-Aligned Multimodal Implicit Field for LiDAR-Camera Joint Synthesis

• Attentive Illumination Decomposition Model for Multi-Illuminant White Balancing

• Misalignment-Robust Frequency Distribution Loss for Image Transformation
  :star:code

• Boosting Neural Representations for Videos with a Conditional Decoder

• SeMoLi: What Moves Together Belongs Together

• VideoMAC: Video Masked Autoencoders Meet ConvNets

• WWW: A Unified Framework for Explaining What Where and Why of Neural Networks by Interpretation of Neuron Concepts

• Integrating Efficient Optimal Transport and Functional Maps For Unsupervised Shape Correspondence Learning

• Training-Free Pretrained Model Merging
  :star:code

• Neural Redshift: Random Networks are not Random Functions

• Ego-Exo4D: Understanding Skilled Human Activity from First- and Third-Person Perspectives

• SCE-MAE: Selective Correspondence Enhancement with Masked Autoencoder for Self-Supervised Landmark Estimation自监督关键点估计

• Unsupervised Feature Learning with Emergent Data-Driven Prototypicality

• LORS: Low-rank Residual Structure for Parameter-Efficient Network Stacking
  :star:code
  :thumbsup:LORS算法：低秩残差结构用于参数高效网络堆叠，参数少、成本低、内存小

• HIMap: HybrId Representation Learning for End-to-end Vectorized HD Map Construction

• Holo-Relighting: Controllable Volumetric Portrait Relighting from a Single Image

• Desigen: A Pipeline for Controllable Design Template Generation
  :star:code

• S2MVTC: a Simple yet Efficient Scalable Multi-View Tensor Clustering
  :star:code

• Semantically-Shifted Incremental Adapter-Tuning is A Continual ViTransformer

• Rewrite the Stars
  :star:code

• Neural Refinement for Absolute Pose Regression with Feature Synthesis
  :star:code
  :house:project

• Instruct 4D-to-4D: Editing 4D Scenes as Pseudo-3D Scenes Using 2D Diffusion

• Localization Is All You Evaluate: Data Leakage in Online Mapping Datasets and How to Fix It
  :star:code

• FairDeDup: Detecting and Mitigating Vision-Language Fairness Disparities in Semantic Dataset Deduplication
  :house:project

• Unsupervised Deep Unrolling Networks for Phase Unwrapping相位展开

• 压缩感知

  • Reconstruction-free Cascaded Adaptive Compressive Sensing
  • UFC-Net: Unrolling Fixed-point Continuous Network for Deep Compressive Sensing
  • CPP-Net: Embracing Multi-Scale Feature Fusion into Deep Unfolding CP-PPA Network for Compressive Sensing

• 数据增强

  • DiffuseMix: Label-Preserving Data Augmentation with Diffusion Models

2020 年论文分类汇总戳这里

↘️CVPR-2020-Papers ↘️ECCV-2020-Papers

2021 年论文分类汇总戳这里

↘️ICCV-2021-Papers ↘️CVPR-2021-Papers

2022 年论文分类汇总戳这里

↘️CVPR-2022-Papers ↘️WACV-2022-Papers ↘️ECCV-2022-Papers

扫码CV君微信(注明：CVPR)入微信交流群：