[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72037":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":27,"readmeContent":28,"aiSummary":29,"trendingCount":16,"starSnapshotCount":16,"syncStatus":17,"lastSyncTime":30,"discoverSource":31},72037,"OOTDiffusion","levihsu\u002FOOTDiffusion","levihsu","[AAAI 2025] Official implementation of \"OOTDiffusion: Outfitting Fusion based Latent Diffusion for Controllable Virtual Try-on\"","",null,"Python",6555,954,78,85,0,2,4,12,6,39.94,"Other",false,"main",true,[],"2026-06-12 02:02:57","# OOTDiffusion\nThis repository is the official implementation of OOTDiffusion\n\n🤗 [Try out OOTDiffusion](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Flevihsu\u002FOOTDiffusion)\n\n(Thanks to [ZeroGPU](https:\u002F\u002Fhuggingface.co\u002Fzero-gpu-explorers) for providing A100 GPUs)\n\n\u003C!-- Or [try our own demo](https:\u002F\u002Footd.ibot.cn\u002F) on RTX 4090 GPUs -->\n\n> **OOTDiffusion: Outfitting Fusion based Latent Diffusion for Controllable Virtual Try-on** [[arXiv paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.01779)]\u003Cbr>\n> [Yuhao Xu](http:\u002F\u002Flevihsu.github.io\u002F), [Tao Gu](https:\u002F\u002Fgithub.com\u002FT-Gu), [Weifeng Chen](https:\u002F\u002Fgithub.com\u002FShineChen1024), [Chengcai Chen](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FChengcai-Chen)\u003Cbr>\n> Xiao-i Research\n\n\nOur model checkpoints trained on [VITON-HD](https:\u002F\u002Fgithub.com\u002Fshadow2496\u002FVITON-HD) (half-body) and [Dress Code](https:\u002F\u002Fgithub.com\u002Faimagelab\u002Fdress-code) (full-body) have been released\n\n* 🤗 [Hugging Face link](https:\u002F\u002Fhuggingface.co\u002Flevihsu\u002FOOTDiffusion) for ***checkpoints*** (ootd, humanparsing, and openpose)\n* 📢📢 We support ONNX for [humanparsing](https:\u002F\u002Fgithub.com\u002FGoGoDuck912\u002FSelf-Correction-Human-Parsing) now. Most environmental issues should have been addressed : )\n* Please also download [clip-vit-large-patch14](https:\u002F\u002Fhuggingface.co\u002Fopenai\u002Fclip-vit-large-patch14) into ***checkpoints*** folder\n* We've only tested our code and models on Linux (Ubuntu 22.04)\n\n![demo](images\u002Fdemo.png) \n![workflow](images\u002Fworkflow.png) \n\n## Installation\n1. Clone the repository\n\n```sh\ngit clone https:\u002F\u002Fgithub.com\u002Flevihsu\u002FOOTDiffusion\n```\n\n2. Create a conda environment and install the required packages\n\n```sh\nconda create -n ootd python==3.10\nconda activate ootd\npip install torch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2\npip install -r requirements.txt\n```\n\n## Inference\n1. Half-body model\n\n```sh\ncd OOTDiffusion\u002Frun\npython run_ootd.py --model_path \u003Cmodel-image-path> --cloth_path \u003Ccloth-image-path> --scale 2.0 --sample 4\n```\n\n2. Full-body model \n\n> Garment category must be paired: 0 = upperbody; 1 = lowerbody; 2 = dress\n\n```sh\ncd OOTDiffusion\u002Frun\npython run_ootd.py --model_path \u003Cmodel-image-path> --cloth_path \u003Ccloth-image-path> --model_type dc --category 2 --scale 2.0 --sample 4\n```\n\n## Citation\n```\n@article{xu2024ootdiffusion,\n  title={OOTDiffusion: Outfitting Fusion based Latent Diffusion for Controllable Virtual Try-on},\n  author={Xu, Yuhao and Gu, Tao and Chen, Weifeng and Chen, Chengcai},\n  journal={arXiv preprint arXiv:2403.01779},\n  year={2024}\n}\n```\n\n## Star History\n\n[![Star History Chart](https:\u002F\u002Fapi.star-history.com\u002Fsvg?repos=levihsu\u002FOOTDiffusion&type=Date)](https:\u002F\u002Fstar-history.com\u002F#levihsu\u002FOOTDiffusion&Date)\n\n## TODO List\n- [x] Paper\n- [x] Gradio demo\n- [x] Inference code\n- [x] Model weights\n- [ ] Training code\n","OOTDiffusion是一个基于潜在扩散模型的可控虚拟试衣项目。该项目通过融合服装与人体图像，实现了高质量的虚拟试穿效果，并支持半身和全身试穿。其核心功能包括使用预训练模型进行推理、支持ONNX格式的人体分割以及灵活调整试穿参数。技术上，OOTDiffusion利用了VITON-HD和Dress Code数据集训练的模型，结合CLIP视觉模型增强语义理解能力。适用于时尚电商、在线服装展示等场景，帮助用户在购买前更直观地看到衣物穿着效果。","2026-06-11 03:40:04","high_star"]