/AnimateZero

Official PyTorch implementation for the paper "AnimateZero: Video Diffusion Models are Zero-Shot Image Animators"

AnimateZero: Video Diffusion Models are Zero-Shot Image Animators

Jiwen Yu1, Xiaodong Cun2, Chenyang Qi3, Yong Zhang2, Xintao Wang2, Ying Shan2, Jian Zhang1

1 Peking University, 2 Tencent AI Lab, 3 HKUST

arXiv Camera Ready Paper Home Page

Gallery

We have showcased some results of image animation using AnimateZero on generated images below.

More results can be found on our Project Page.

Image 1 Animation 1 Image 2 Animation 1 Image 3 Animation 1
"1girl, underwater, swimsuit, ..." "1girl, black jacket, long sleeves, ..." "dark fantasy, purple eyes, ..."
Image 1 Animation 1 Image 2 Animation 1 Image 3 Animation 1
"school uniform, JK, sketch, ..." "pixel art, cat ears, blonde hair, ..." "a cat head, look to one side"
Image 1 Animation 1 Image 2 Animation 1 Image 3 Animation 1
"waves hit the beach" "freckles, orange hair, glasses, ..." "lighter, flame, candle"

BibTeX

@article{yu2023animatezero,
  title={AnimateZero: Video Diffusion Models are Zero-Shot Image Animators},
  author={Yu, Jiwen and Cun, Xiaodong and Qi, Chenyang and Zhang, Yong and Wang, Xintao and Shan, Ying and Zhang, Jian},
  journal={arXiv preprint arXiv:2312.03793},
  year={2023}
}