JSHZT

[Question] The magnitude and change of loss convergence during the training process.

2

Excellent work! your work inspires me. Now I am trying to train this model. I am confused during the rough training stage. The calculation of identity loss has been maintained...

get_landmarks_from_batch returns an empty list

1

My code is as follows: ``` imgs = imgs.permute(0, 3, 1, 2)# B x C x H x W landmark = self.face_algm.get_landmarks_from_batch(imgs) ``` The picture I used is the frame...

Excellent work! Using vocaset and our own training set based on FLAME, I got better results. But what confuses me is the training speed. If I use your code, the...

Running the demo code seems to require the v1.3 model, now updated to v1.6, how should I get the v1.3 model?The new version of the model doesn't seem to work...

I saw that the data contained in this download method only has a few identities, so I doubt whether it is an incomplete Ted Gesture data set. If it is...

Poor performance when using videos from other datasets

2

I used the video from vgg-sound, I used the default question, but I found that the answer from minigpt4-video has nothing to do with the video.

Problems with rendering

2

Can this work be used to render **BFM face models**? I get this warning when I try to render this model:**Bin size was too small in the coarse rasterization phase....