🎉 CVPR 2025 Oral Paper List

A collection of 95 oral papers from CVPR 2025, organized by topic, with links to papers, project pages, and code.

Official resources: Program · Proceedings · Awards

📚 Browse by Topic

Topic	Papers
3D Geometry & Reconstruction	13
Rendering & 3D/4D Generation	10
Humans, Avatars & Motion	5
Image & Video Generation	15
Vision-Language & Video Understanding	17
Embodied AI & Autonomous Driving	6
Segmentation & Detection	5
Computational Imaging & Restoration	8
Learning, Efficiency & Trustworthiness	13
Medical & Scientific Vision	3

🏆 Awards

Award labels follow the official CVPR 2025 results.

Award	Paper
Best Paper	VGGT
Best Student Paper	Neural Inverse Rendering from Propagating Light
Best Paper Honorable Mention	MegaSaM
Best Paper Honorable Mention	Navigation World Models
Best Paper Honorable Mention	Molmo and PixMo
Best Paper Honorable Mention	3D Student Splatting and Scooping
Best Student Paper Honorable Mention	Generative Multimodal Pretraining with Discrete Diffusion Timestep Tokens

📐 3D Geometry & Reconstruction

Paper	Links
VGGT: Visual Geometry Grounded Transformer	Paper · Project · Code
CUT3R — Continuous 3D Perception Model with Persistent State	Paper · Project · Code
MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision	Paper · Project · Code
FoundationStereo: Zero-Shot Stereo Matching	Paper · Project · Code
Multi-view Reconstruction via SfM-guided Monocular Depth Estimation	Paper · Project · Code
MV-DUSt3R+: Single-Stage Scene Reconstruction from Sparse Views In 2 Seconds	Paper · Project · Code
MegaSaM: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos	Paper · Project · Code
Stereo4D: Learning How Things Move in 3D from Internet Stereo Videos	Paper · Project · Code
Zero-Shot Monocular Scene Flow Estimation in the Wild	Paper · Project · Code
DIFIX3D+: Improving 3D Reconstructions with Single-Step Diffusion Models	Paper · Project · Code
TacoDepth: Towards Efficient Radar-Camera Depth Estimation with One-stage Fusion	Paper · Code
Convex Relaxation for Robust Vanishing Point Estimation in Manhattan World	Paper · Code
Camera Resection from Known Line Pencils and a Radially Distorted Scanline	Paper · Code

✨ Rendering & 3D/4D Generation

Paper	Links
Neural Inverse Rendering from Propagating Light	Paper · Project · Code
Diffusion Renderer: Neural Inverse and Forward Rendering with Video Diffusion Models	Paper · Project · Code
3D Student Splatting and Scooping	Paper · Code
3DGUT: Enabling Distorted Cameras and Secondary Rays in Gaussian Splatting	Paper · Code
Time of the Flight of the Gaussians: Optimizing Depth Indirectly in Dynamic Radiance Fields	Paper · Project · Code
FluidNexus: 3D Fluid Reconstruction and Prediction from a Single Video	Paper · Project · Code
CraftsMan3D: High-fidelity Mesh Generation with 3D Native Diffusion and Interactive Geometry Refiner	Paper · Project · Code
CAT4D: Create Anything in 4D with Multi-View Video Diffusion Models	Paper · Project
DNF: Unconditional 4D Generation with Dictionary-based Neural Fields	Paper · Project · Code
Birth and Death of a Rose	Paper · Project

🧍 Humans, Avatars & Motion

Paper	Links
CAP4D: Creating Animatable 4D Portrait Avatars with Morphable Multi-View Diffusion Models	Paper · Project · Code
Reconstructing Humans with a Biomechanically Accurate Skeleton	Paper · Project · Code
MEGA: Masked Generative Autoencoder for Human Mesh Recovery	Paper
TokenHSI: Unified Synthesis of Physical Human-Scene Interactions through Task Tokenization	Paper · Project · Code
EgoLM: Multi-Modal Language Model of Egocentric Motions	Paper · Project

🎨 Image & Video Generation

Paper	Links
Infinity∞: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis	Paper · Project · Code
RandAR: Decoder-only Autoregressive Visual Generation in Random Orders	Paper · Project · Code
Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Models	Paper · Code
Alias-Free Latent Diffusion Models: Improving Fractional Shift Equivariance of Diffusion Latent Space	Paper · Project · Code
Autoregressive Distillation of Diffusion Transformers	Paper · Code
Language-Guided Image Tokenization for Generation	Paper · Project
AnyEdit: Mastering Unified High-Quality Image Editing for Any Idea	Paper · Project · Code
CustAny: Customizing Anything from A Single Example	Paper · Project · Code
DreamRelation: Bridging Customization and Relation Generation	Paper · Project · Code
Minority-Focused Text-to-Image Generation via Prompt Optimization	Paper · Code
DesignDiffusion: High-Quality Text-to-Design Image Generation with Diffusion Models	Paper
Motion Prompting: Controlling Video Generation with Motion Trajectories	Paper · Project
Go-with-the-Flow: Motion-Controllable Video Diffusion Models Using Real-Time Warped Noise	Paper · Project · Code
LookingGlass: Generative Anamorphoses via Laplacian Pyramid Warping	Paper · Project
Reanimating Images using Neural Representations of Dynamic Stimuli	Paper

💬 Vision-Language & Video Understanding

Paper	Links
Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models	Paper · Blog · Code
Generative Multimodal Pretraining with Discrete Diffusion Timestep Tokens	Paper · Project · Code
LoRASculpt: Sculpting LoRA for Harmonizing General and Specialized Knowledge in Multimodal Large Language Models	Paper
OPA-DPO — Mitigating Hallucinations in Large Vision-Language Models via DPO: On-Policy Data Hold the Key	Paper · Project · Code
Seeing Far and Clearly: Mitigating Hallucinations in MLLMs with Attention Causal Decoding	Paper · Project · Code
Identifying and Mitigating Position Bias of Multi-image Vision-Language Models	Paper
Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces	Paper · Project · Code
Video-XL: Extra-Long Vision Language Model for Hour-Scale Video Understanding	Paper · Code
VideoEspresso: A Large-Scale Chain-of-Thought Dataset for Fine-Grained Video Reasoning via Core Frame Selection	Paper · Code
OpenING: A Comprehensive Benchmark for Judging Open-ended Interleaved Image-Text Generation	Paper · Project · Code
Q-Eval-100K: Evaluating Visual Quality and Alignment Level for Text-to-Vision Content	Paper · Code
SEAL: Semantic Attention Learning for Long Video Representation	Paper · Code
Learning Audio-guided Video Representation with Gated Attention for Video-Text Retrieval	Paper
Temporal Alignment-Free Video Matching for Few-shot Action Recognition	Paper · Code
Viewpoint Rosetta Stone: Unlocking Unpaired Ego-Exo Videos for View-invariant Representation Learning	Paper · Project
Temporally Consistent Object-Centric Learning by Contrasting Slots	Paper · Project · Code
The PanAf-FGBG Dataset: Understanding the Impact of Backgrounds in Wildlife Behaviour Recognition	Paper · Project

🤖 Embodied AI & Autonomous Driving

Paper	Links
Navigation World Models	Paper · Project · Code
RoboSpatial: Teaching Spatial Understanding to 2D and 3D Vision-Language Models for Robotics	Paper · Project · Code
From Multimodal LLMs to Generalist Embodied Agents: Methods and Lessons	Paper
PDFactor: Learning Tri-Perspective View Policy Diffusion Field for Multi-Task Robotic Manipulation	Paper
GROVE: A Generalized Reward for Learning Open-Vocabulary Physical Skill	Paper · Project · Code
Closed-Loop Supervised Fine-Tuning of Tokenized Traffic Models	Paper · Project · Code

🎯 Segmentation & Detection

Paper	Links
SegEarth-OV: Towards Training-Free Open-Vocabulary Segmentation for Remote Sensing Images	Paper · Project · Code
Effective SAM Combination for Open-Vocabulary Semantic Segmentation	Paper
Keep the Balance: A Parameter-Efficient Symmetrical Framework for RGB+X Semantic Segmentation	Paper
Towards Explicit Geometry-Reflectance Collaboration for Generalized LiDAR Segmentation in Adverse Weather	Paper
Efficient Test-time Adaptive Object Detection via Sensitivity-Guided Pruning	Paper

📷 Computational Imaging & Restoration

Paper	Links
Learned Binocular-Encoding Optics for RGBD Imaging Using Joint Stereo and Focus Cues	Paper · Project · Code
Opportunistic Single-Photon Time of Flight	Paper
Descriptor-In-Pixel : Point-Feature Tracking For Pixel Processor Arrays	Paper · Project
Removing Reflections from RAW Photos	Paper · Project
DORNet: A Degradation Oriented and Regularized Network for Blind Depth Super-Resolution	Paper · Code
Improving Diffusion Inverse Problem Solving with Decoupled Noise Annealing	Paper · Project · Code
DiffFNO: Diffusion Fourier Neural Operator	Paper · Project
Semi-Supervised State-Space Model with Dynamic Stacking Filter for Real-World Video Deraining	Paper

🧠 Learning, Efficiency & Trustworthiness

Paper	Links
OverLoCK: An Overview-first-Look-Closely-next ConvNet with Context-Mixing Dynamic Kernels	Paper · Code
CleanDIFT: Diffusion Features without Noise	Paper · Project · Code
LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions	Paper · Code
Do We Always Need the Simplicity Bias? Looking for Optimal Inductive Biases in the Wild	Paper
Rethinking Spiking Self-Attention Mechanism: Implementing α-XNOR Similarity Calculation in Spiking Transformers	Paper
Towards Universal Dataset Distillation via Task-Driven Diffusion	Paper
Gromov–Wasserstein Problem with Cyclic Symmetry	Paper
UniAP: Unifying Inter- and Intra-Layer Automatic Parallelism by Mixed Integer Quadratic Programming	Paper
Enhancing Diversity for Data-free Quantization	Paper
Geometric Knowledge-Guided Localized Global Distribution Alignment for Federated Learning	Paper · Code
Black-Box Forgery Attacks on Semantic Watermarks for Diffusion Models	Paper · Code
Adv-CPG: A Customized Portrait Generation Framework with Facial Adversarial Attacks	Paper · Code
Rethinking Vision-Language Model in Face Forensics: Multi-Modal Interpretable Forged Face Detector	Paper · Code

🔬 Medical & Scientific Vision

Paper	Links
TopoCellGen: Generating Histopathology Cell Topology with a Diffusion Model	Paper · Code
Enhancing SAM with Efficient Prompting and Preference Optimization for Semi-supervised Medical Image Segmentation	Paper
IceDiff: High Resolution and High-Quality Arctic Sea Ice Forecasting with Generative Diffusion Prior	Paper

📎 Related Reading

4 additional papers from the original collection

These entries are retained for reference and are not counted among the 95 papers in the official oral program.

Paper	Links
Exploring CLIP's Dense Knowledge for Weakly Supervised Semantic Segmentation	Paper · Code
FedSPA: Generalizable Federated Graph Learning under Homophily Heterogeneity	Paper
One Category One Prompt: Dataset Distillation using Diffusion Models	Paper · Code
Video-XL-Pro: Reconstructive Token Compression for Extremely Long Video Understanding	Paper · Code

🌷 Contributing & Acknowledgments

Missing a link or spotted a mistake? Open an issue or submit a pull request with the paper title and updated links.

Thanks to the paper authors for sharing their work, and to cvpr25_oral_gpu_info for the original collection reference.