Views
No views yet
git clone https://github.com/2y7c3/3DV-TON.git
cd 3DV-TON
pip install -r requirements.txt
cd preprocess/model/DensePose/detectron2/projects/DensePose
pip install -e .
## install GVHMR
## see https://github.com/zju3dv/GVHMR/blob/main/docs/INSTALL.md
## replace GVHMR/hmr4d/utils/vis/renderer.py with our preprocess/renderer.py./preprocess/ckpts. Then you can use our cloth masker to generate agnostic mask videos for improved try-on results../demos/ — run the following commands to test them.python infer.py --config ./configs/inference/demo_test.yaml1# 1. generate agnostic mask (type: 'upper', 'lower', 'overall')
2cd preprocess
3python seg_mask.py --input demos/videos/video.mp4 --output demos/ --type overall
4
5# 2. use GVHMR to generate SMPL video
6
7# 3. use image tryon model to generate tryon image (e.g. CaTVTON)
8
9# 4. generate textured 3d mesh
10
11# 5. modify demo_test.yaml, then run
12python infer.py --config ./configs/inference/demo_test.yaml1@article{wei20253dv,
2 title={3dv-ton: Textured 3d-guided consistent video try-on via diffusion models},
3 author={Wei, Min and Yu, Chaohui and Zhou, Jingkai and Wang, Fan},
4 journal={arXiv preprint arXiv:2504.17414},
5 year={2025}
6}