git clone https://github.com/yisol/IDM-VTON.git
cd IDM-VTON

conda env create -f environment.yaml
conda activate idm

train
|-- image
|-- image-densepose
|-- agnostic-mask
|-- cloth
|-- vitonhd_train_tagged.json

test
|-- image
|-- image-densepose
|-- agnostic-mask
|-- cloth
|-- vitonhd_test_tagged.json

DressCode
|-- dresses
    |-- images
    |-- image-densepose
    |-- dc_caption.txt
    |-- ...
|-- lower_body
    |-- images
    |-- image-densepose
    |-- dc_caption.txt
    |-- ...
|-- upper_body
    |-- images
    |-- image-densepose
    |-- dc_caption.txt
    |-- ...

accelerate launch train_xl.py \
    --gradient_checkpointing --use_8bit_adam \
    --output_dir=result --train_batch_size=6 \
    --data_dir=DATA_DIR

accelerate launch inference.py \
    --width 768 --height 1024 --num_inference_steps 30 \
    --output_dir "result" \
    --unpaired \
    --data_dir "DATA_DIR" \
    --seed 42 \
    --test_batch_size 2 \
    --guidance_scale 2.0

accelerate launch inference_dc.py \
    --width 768 --height 1024 --num_inference_steps 30 \
    --output_dir "result" \
    --unpaired \
    --data_dir "DATA_DIR" \
    --seed 42 
    --test_batch_size 2
    --guidance_scale 2.0
    --category "upper_body" 

ckpt
|-- densepose
    |-- model_final_162be9.pkl
|-- humanparsing
    |-- parsing_atr.onnx
    |-- parsing_lip.onnx

|-- openpose
    |-- ckpts
        |-- body_pose_model.pth
    

@article{choi2024improving,
  title={Improving Diffusion Models for Authentic Virtual Try-on in the Wild},
  author={Choi, Yisol and Kwak, Sangkyung and Lee, Kyungmin and Choi, Hyungwon and Shin, Jinwoo},
  journal={arXiv preprint arXiv:2403.05139},
  year={2024}
}