dreadlord1984 / viton Goto Github PK

Code and dataset for the CVPR 2018 paper "VITON: An Image-based Virtual Try-on Network"

The person representation used in this paper are extracted by a 2D pose estimator and a human parser:

• Realtime Multi-Person Pose Estimation
• Self-supervised Structure-sensitive Learning

The dataset is no longer publicly available due to copyright issues. For thoese who have already downloaded the dataset, please note that using or distributing it is illegal!

Download pretrained models on Google Drive. Put them under model/ folder.

Run test_stage1.sh to do the inference. The results are in results/stage1/images/. results/stage1/index.html visualizes the results.

Run the matlab script shape_context_warp.m to extract the TPS transformation control points.

Then test_stage2.sh will do the refinement and generate the final results, which locates in results/stage2/images/. results/stage2/index.html visualizes the results.

Go inside prepare_data.

First run extract_tps.m. This will take sometime, you can try run it in parallel or directly download the pre-computed TPS control points via Google Drive and put them in data/tps/.

Then run ./preprocess_viton.sh, and the generated TF records will be in prepare_data/tfrecord.

Run train_stage1.sh

Run train_stage2.sh

If this code or dataset helps your research, please cite our paper:

@inproceedings{han2017viton,
  title = {VITON: An Image-based Virtual Try-on Network},
  author = {Han, Xintong and Wu, Zuxuan and Wu, Zhe and Yu, Ruichi and Davis, Larry S},
  booktitle = {CVPR},
  year  = {2018},
}