Siziff
Siziff
HI! @SteveJunGao 1. I read the article very carefully and I can't understand, the blender for training generates 24 views from different sides of the 3d figure, but where are...
@ajhamdi hi) I use the blender for this, there you can import a file in .obj format and export it in .usd format.
same
SOOOO? Has this problem been solved? (pytorch == 2.1.2+cu121, CUDA == 12.2 , GPU only one v100 16 GB )
@ltkong218 I installed the same versions as yours and it worked. Thanks :)
@ltkong218 I have seen this solution, but it does not describe the conversion process, but only gives the already converted models :(
@xwqok You can use the open source program MeshLab for this . There you will be able to output two sys files in different colors in the same space.
@paulovasconcellos-hotmart Was it possible to solve this problem? When can I expect the updated code??
> [@yaleimeng](https://github.com/yaleimeng)But I tried SyncTalk, I used a 5-minute HD video, and the trained lip movements were 99% correct, but the facial expressions were not as natural as real people....
@gaurangbharti1 Have you been able to achieve real-time model inference in your project?