G-force78

RuntimeError: Detected that PyTorch and torchvision were compiled with different CUDA versions. PyTorch has CUDA Version=11.7 and torchvision has CUDA Versio=11.6. Please reinstall the torchvision that matches your PyTorch install.

12

When launching training Seems to be an error everywhere with this so not specific to this repo. Any ideas how to fix?

Adapt to use SDD-1B?

1

https://huggingface.co/segmind/SSD-1B or https://github.com/luosiallen/latent-consistency-model

The size of tensor a (4) must match the size of tensor b (8)

2

Using these arguments !python3 /content/control-a-video/inference.py --prompt "a bear practicing kungfu, with a background of mountains" --input_video /content/kungfubear.mp4 --control_mode depth --num_sample_frames 24 --inference_step 10 --guidance_scale 5 --init_noise_thres 0.75 FPS 8 output...

How to use emo_talk blenshape capture with this implementation?

1

As title says

I'm hoping some wizz can adapt the long text script to add an option to read from txt file. Although currently it does seem to do quite well with longer...

Video to video?

1

Any plans to release a similar implementation to liveportrait? I think it uses boundary boxes to keep it focussed on the face

As described, would be good to be able to add blinking and the frequency of it