metadata
license: other
license_name: multiple-licenses
license_link: https://huggingface.co/AntResearch/4DAnyone/blob/main/LICENSE.md
pipeline_tag: video-to-video
tags:
- video-generation
- multiview-video-generation
- 4D-human-reconstruction
- novel-view-synthesis
4DAnyone: Create Anyone in 4D from a Casual Monocular Video
Project Page | Paper | Code
4DAnyone turns a casual monocular video into multi-view videos, enabling downstream 4DGS reconstruction.
Models
4danyone/
├── model.safetensors
├── smplx_to_goliath70.pt
├── Wan2.2_VAE.pth
├── prompt_context.safetensors
└── Wan22_TI2V_5B_Turbo_lora_rank_64_fp16.safetensors
gvhmr/
├── gvhmr_siga24_release.ckpt
├── epoch=10-step=25000.ckpt
├── vitpose-h-multi-coco.pth
└── yolov8x.pt
perceptual/
└── imagenet-vgg-verydeep-19-conv.safetensors
data/source/pexels/
└── 20 curated 121-frame example videos
Licensing is asset-specific. See LICENSE for the exact scope and licenses/ for the applicable terms and notices.
Citation
If you find 4DAnyone useful or interesting, please cite our work and give this model a ❤️ like:
@article{jin2026fdanyone,
title={4DAnyone: Create Anyone in 4D from a Casual Monocular Video},
author={Jin, Yudong and Xie, Tao and Zhang, Qihang and Shen, Zehong and Xu, Zhen and Shen, Yujun and Bao, Hujun and Zhou, Xiaowei and Xu, Yinghao},
journal={arXiv preprint arXiv:2608.20335},
year={2026},
url={https://arxiv.org/abs/2608.20335}
}