UniAnimate
UniAnimate copied to clipboard
Code for SCIS-2025 Paper "UniAnimate: Taming Unified Video Diffusion Models for Consistent Human Image Animation".
Thanks for your interesting work! I have tried to infer a 75 frames video on A100 in 512*768, which will take about 3mins. I also tried to use more cards,...
Hi, thanks for your great work! UniAnimate is the best work I know for human image animation! I have two questions regarding the `random_ref_frame`, `vit_frame`, and `local_image`: 1. In unet_unianimate.py,...
Hi,I try to finetune the model on my own dataset, but I am not very familiar with the diffusion model's trainning. I would like to ask where I can get...
Thank you very much for open sourcing such excellent work. I have two questions I would like to ask: 1. I would like to know if you have compared the...
有些效果还是挺好的,不过项目测试后发现 参数不能设置输出目录,最后保存视频的API设计得比较复杂,没法只保存单列不拼接的视频,不方便制作gradio界面 (AIFSH的ComfyUI插件好像烂尾了[“😂”]), 另外输出目录生成了两个视频,不知道是什么,没看出有什么区别, 如果发布了性能更好的模型,希望改进一下代码接口
CUDA error (third_party/flash-attention/csrc/flash_attn/src/fmha_fwd_launch_template.h:89): no kernel image is available for execution on the device
Hi, I want to ask if there are any ways to improve the quality of videos generated by UniAnimate (such as using super-resolution post-processing, since I noticed that tft2v offers...
I always find that I am missing many things in the requirements, in this case also that it uses nccl, for multiple GPUs, which is not yet available in Windows,...
键重复了,去掉一个吧  ``` ConstructorError: while constructing a mapping in "/root/UniAnimate/configs/UniAnimate_infer_long.yaml", line 3, column 1 found duplicate key use_fp16 ```