<p align="center">

<h2 align="center">Animate-X：基于增强运动表征的通用角色图像动画</h2> <p align="center"> <a href=""><strong>谭帅</strong></a> · <a href="https://scholar.google.com/citations?user=BwdpTiQAAAAJ"><strong>龚标</strong></a>^† · <a href="https://scholar.google.com/citations?user=cQbXvkcAAAAJ"><strong>王翔</strong></a> · <a href="https://scholar.google.com/citations?user=ZO3OQ-8AAAAJ"><strong>张世伟</strong></a> <br> <a href="https://openreview.net/profile?id=~DanDan_Zheng1"><strong>郑丹丹</strong></a> · <a href="https://scholar.google.com.hk/citations?user=S8FmqTUAAAAJ"><strong>郑若冰</strong></a> · <a href="https://scholar.google.com/citations?user=hMDQifQAAAAJ"><strong>郑克成</strong></a> · <a href="https://openreview.net/profile?id=~Jingdong_Chen1"><strong>陈京东</strong></a> · <a href="https://openreview.net/profile?id=~Ming_Yang2"><strong>杨明</strong></a>
<br> <br> <a href="https://arxiv.org/abs/2410.10306"><img src='https://img.shields.io/badge/arXiv-Animate--X-red' alt='论文 PDF'></a> <a href='https://lucaria-academy.github.io/Animate-X/'><img src='https://img.shields.io/badge/Project_Page-Animate--X-blue' alt='项目主页'></a> <a href='https://mp.weixin.qq.com/s/vDR4kPLqnCUwfPiBNKKV9A'><img src='https://badges.aleen42.com/src/wechat.svg'></a> <a href='https://github.com/antgroup/animate-x'><img src='https://img.shields.io/badge/Code-Animate--X-yellow'></a> <br> <b></a>蚂蚁集团   |   </a>通义实验室 </b> <br> </p> </p>

本仓库包含 Animate-X 的模型权重文件：

• "checkpoints/dw-ll_ucoco_384.onnx"：dwpose 提取的模型权重。

• "checkpoints/open_clip_pytorch_model.bin"：clip 嵌入的模型权重。

• "checkpoints/animate-x_ckpt.pth"：Animate-X 中 X-角色图像动画的模型权重（32 帧）。

• "checkpoints/yolox_l.onnx"：dwpose 提取的模型权重。

• "checkpoints/v2-1_512-ema-pruned.ckpt"：Stable Diffusion 的模型权重。

BibTeX

如果本仓库对您有帮助，请引用我们的相关技术论文。

@article{AnimateX2025,
  title={Animate-X: Universal Character Image Animation with Enhanced Motion Representation},
  author={Tan, Shuai and Gong, Biao and Wang, Xiang and Zhang, Shiwei and Zheng, Dandan and Zheng, Ruobing and Zheng, Kecheng and Chen, Jingdong and Yang, Ming},
  journal={arXiv preprint arXiv:2410.10306},
  year={2025}
}