Licheng Zhong · Lixin Yang · Kailin Li · Haoyu Zhen · Mei Han · Cewu Lu
Project Page | arXiv | Data
demo_video.mp4
git clone https://github.com/Colmar-zlicheng/Color-NeuS.git
cd Color-NeuS
conda install pytorch torchvision torchaudio pytorch-cuda=11.8 -c pytorch -c nvidia
pip install -r requirements.txt
pip install "git+https://github.com/facebookresearch/pytorch3d.git"
- IHO Video
- DTU (NeuS Preprocess Link | Raw Link)
- BlendedMVS (NeuS Preprocess Link | Raw Link)
- OmniObject3D
- set
${DATASET}
as one in[iho, dtu, bmvs, omniobject3d]
- set
${OBJECT_NAME}
as the name of the object in the dataset
python train.py -g 0 --cfg config/Color_NeuS_${DATASET}.yml -obj ${OBJECT_NAME} --exp_id ${EXP_ID}
-g, --gpu_id
, visible GPUs for training, e.g.-g 0
. Only supports single GPU.--exp_id
specify the name of experiment, e.g.--exp_id ${EXP_ID}
. When--exp_id
is provided, the code requires that no uncommitted change is remained in the git repo. Otherwise, it defaults to'default'
for training and'eval_{cfg}_{OBJECT_NAME}'
for evaluation. All results will be saved inexp/${EXP_ID}*{timestamp}
.
# IHO Video: ghost_bear
python train.py -g 0 --cfg config/Color_NeuS_iho.yml -obj ghost_bear --exp_id Color_NeuS_iho_ghost_bear
# DTU: dtu_scan83
python train.py -g 0 --cfg config/Color_NeuS_dtu.yml -obj 83 --exp_id Color_NeuS_dtu_83
# BlendedMVS: bmvs_bear
python train.py -g 0 --cfg config/Color_NeuS_bmvs.yml -obj bear --exp_id Color_NeuS_bmvs_bear
# OmniObject3D: doll_002
python train.py -g 0 --cfg config/Color_NeuS_omniobject3d.yml -obj doll_002 --exp_id Color_NeuS_omniobject3d_doll_002
All the training checkpoints are saved at exp/${EXP_ID}_{timestamp}/checkpoints/
We also provide our implementation of NeuS in this repo. To train NeuS, you can replace Color_NeuS_${DATASET}.yml
with NeuS_${DATASET}.yml
in the above command line, such as:
# IHO Video: ghost_bear
python train.py -g 0 --cfg config/NeuS_iho.yml -obj ghost_bear --exp_id NeuS_iho_ghost_bear
- set corresponding
${DATASET}
and${OBJECT_NAME}
as above - set
${PATH_TO_CHECKPOINT}
as the path to the checkpoint (NeuS_Trainer.pth.tar) to be loaded
python evaluation.py -g 0 --cfg config/Color_NeuS_${DATASET}.yml -obj ${OBJECT_NAME} -rr 512 --reload ${PATH_TO_CHECKPOINT}
-rr, --recon_res
is the resolution of the reconstructed mesh. The default value is 512.
The code provided herein are available for usage as specified in the LICENSE file. By downloading and using the code you agree to the terms in the LICENSE.
@inproceedings{zhong2024colorneus,
title = {Color-NeuS: Reconstructing Neural Implicit Surfaces with Color},
author = {Zhong, Licheng and Yang, Lixin and Li, Kailin and Zhen, Haoyu and Han, Mei and Lu, Cewu},
booktitle = {International Conference on 3D Vision (3DV)},
year = {2024}
}
For more questions, please contact Licheng Zhong: [email protected]