GhiXu/Geo-Neus
GhiXu/Geo-Neus has 313 GitHub stars and 19 forks. Star changes are not measured yet. Last checked 2026-11-18 07:44 UTC. Last push 2022-11-29.
Progression mesurée des étoiles
- in 24h
- Mise à jour retardée
- in 7 days
- Mise à jour retardée
- in 15 days
- Mise à jour retardée
- in 1 month
- Mise à jour retardée
Variations nettes sur des périodes qui se chevauchent, pas des totaux quotidiens. Les mesures absentes restent vides. Les étoiles indiquent l’intérêt, pas la qualité.
À propos
Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction (NeurIPS 2022)
ApprentissageREADME.md
OriginalGeo-Neus


This is the official repo for the implementation of Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction, Qiancheng Fu*, Qingshan Xu*, Yew-Soon Ong, Wenbing Tao (* Equal Contribution), NeurIPS 2022.
We will release our pretrained models soon!
Setup
Installation
This code is built with pytorch 1.11.0 and pytorch3d 0.6.2. In addition, other packages listed in requirements.txt are required.
You can create an anaconda environment called geoneus with the required dependencies by running:
conda create -n geoneus python=3.7
conda activate geoneus
conda install pytorch==1.11.0 torchvision==0.12.0 cudatoolkit=11.3 -c pytorch
conda install fvcore iopath
conda install -c bottler nvidiacub
conda install pytorch3d -c pytorch3d
pip install -r requirements.txt
Data
Download the DTU data from Google Drive, which provides the files on sparse points and image pairs.
Running
- Training
python exp_runner.py --mode train --conf ./confs/womask.conf --case <case_name>
- Extract surface from trained model
python exp_runner.py --mode validate_mesh --conf ./confs/womask.conf --case <case_name> --is_continue
- Evaluation
python eval.py --conf ./confs/womask.conf --case <case_name>
Citation
If you find our work useful in your research, please consider citing:
@article{Fu2022GeoNeus,
title={Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction},
author={Fu, Qiancheng and Xu, Qingshan and Ong, Yew-Soon and Tao, Wenbing},
journal={Advances in Neural Information Processing Systems (NeurIPS)},
year={2022}
}
Acknowlegement
Our code is partially based on NeuS project and some code snippets are borrowed from NeuralWarp. Thanks for these great projects.