Note We have reproduced the other pipeline
Note The image is generated by the model with 50% training progress
This repository is copy from the official implementation of OOTDiffusion
๐ค Try out OOTDiffusion (Thanks to ZeroGPU for providing A100 GPUs)
Or try our own demo on RTX 4090 GPUs
OOTDiffusion: Outfitting Fusion based Latent Diffusion for Controllable Virtual Try-on [arXiv paper]
Yuhao Xu, Tao Gu, Weifeng Chen, Chengcai Chen
Xiao-i Research
Our model checkpoints trained on VITON-HD (half-body) and Dress Code (full-body) have been released
- ๐ค Hugging Face link
- ๐ข๐ข We support ONNX for humanparsing now. Most environmental issues should have been addressed : )
- Please download clip-vit-large-patch14 into checkpoints folder
- We've only tested our code and models on Linux (Ubuntu 22.04)
- Clone the repository
git clone https://github.com/levihsu/OOTDiffusion
- Create a conda environment and install the required packages
conda create -n ootd python==3.10
conda activate ootd
pip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2
pip install -r requirements.txt
- Half-body model
cd OOTDiffusion/run
python run_ootd.py --model_path <model-image-path> --cloth_path <cloth-image-path> --scale 2.0 --sample 4
- Full-body model
Garment category must be paired: 0 = upperbody; 1 = lowerbody; 2 = dress
cd OOTDiffusion/run
python run_ootd.py --model_path <model-image-path> --cloth_path <cloth-image-path> --model_type dc --category 2 --scale 2.0 --sample 4
accelerate launch ootd_train.py --load_height 512 --load_width 384 --dataset_list 'train_pairs.txt' --dataset_mode 'train' --batch_size 16 --train_batch_size 16 --num_train_epochs 200
@article{xu2024ootdiffusion,
title={OOTDiffusion: Outfitting Fusion based Latent Diffusion for Controllable Virtual Try-on},
author={Xu, Yuhao and Gu, Tao and Chen, Weifeng and Chen, Chengcai},
journal={arXiv preprint arXiv:2403.01779},
year={2024}
}
- Paper
- Gradio demo
- Inference code
- Model weights
- Training code
- Distributed and Parallel Training code