somuchtome
somuchtome
> > @jasonborn0 You have to make sure your `GroundingDINO` is installed with a GPU instead of CPU. The groundingdino._C isn't installed without GPU. And the default hyperparameter is `--device...
You can try pip install diffusers==0.10.0 rather than 0.3.0
Thank you! That's right.
Sorry for the mistake. I have updated the link in ReadME.
I have removed the ckpt to Google Drive and updated the ReadME.md. You can download it from the link and contact me if there is any problem.
Yes，I will provide the results in technical report and the link is already added to ReadMe.md
I didn't meet the bug but the face-alignment may not be installed correctly. You can try 'pip install face-alignment==1.4.1' again.
> We encourage the community to submit a PR on SVD implementation of AnyV2V :) Hi~ Is there AnyV2V solution based on "animatediff"? I am not sure which layers' features...
> 请问二阶段加载animatediff的motion model进行finetune的时候，训练刚开始时模型去噪的结果是接近于噪音吗？我训练时是这个结果。但是如果不加载mm的预训练权重，由于零卷积的存在，训练刚开始时模型去噪的结果是正确的人像（也就是一阶段训练出的结果） > > When loading animatediff's motion model for finetune in the second stage, is the model denoising result close to noise at the beginning of training? This...
> Hey @xuanxu92, thanks for your interest in our project. > > Our training code is based on [PIA](https://github.com/open-mmlab/PIA). We re-implement the attention operation of Motion Module and use the...