zwpro/boomvvt

BooM-VVT: Boosting Mask-Free Video Virtual Try-On with Image-Level Pseudo Data

Get the weights

hf download zwpro/boomvvt --local-dir ./ckpt

Inference

python inference.py \
  --video path/to/person.mp4 \
  --cloth path/to/garment.jpg \
  --prompt_path path/to/prompt.txt \
  --garment_type upper_body \
  --tryon_lora_path ./ckpt/qwen-mv-tryon-lora/lora.safetensors \
  --appearance_lora_path ./ckpt/wan-animate-lora/appearance-transfer_lora.safetensors \
  --location_lora_path ./ckpt/wan-animate-lora/localization_lora.safetensors \
  --output ./asset/Output

See the code repo for backbone paths and the full inference CLI.

License

These weights are released under CC BY-NC-SA 4.0. Non-commercial use only.

Citation

@inproceedings{zhang2026boomvvt,
  title     = {BooM-VVT: Boosting Mask-Free Video Virtual Try-On with Image-Level Pseudo Data},
  author    = {Zhang, Wei and Li, Xin and Shi, Peishu and Gao, Jialin and
               Peng, Xuekang and Lian, Zhichao and Jin, Yeying},
  booktitle = {Proceedings of the 34th ACM International Conference on Multimedia},
  year      = {2026},
  doi       = {10.1145/3767308.3835712},
  eprint    = {2609.04120},
  archivePrefix = {arXiv}
}