Jianzhu Guo
Jianzhu Guo
You may try updating your matplotlib or referring to g2ragrawal. matplotlib is just for visualization, you can also use cv(opencv) to visualize it.
Sorry for that I can not re-implement your status : (
This part may be released in the near future.
Sorry for that I may have not enough time to maintain a QQ group.
The data pre-processing part is not ready for release, but you can build the data with your needed pca vector size by processing 300W-LP.
Your compiling env may not be ready. Besides, refer https://github.com/cleardusk/3DDFA_V2/issues/12#issuecomment-697479173
At that time, the inference speed of TF is faster than PyTorch. Now, ONNX is faster. I think there is no need to test the code on TF now?
Compared to the original bfm, `bfm_noneck_v3` clips the shape/exp params, removes the ear/neck region. You can find the mapping across the vertices of the mean shape.
Thanks for your detailed reading, I check it out, you are right. It should be \sqrt{N}, not \sqrt{N/3}. I just made a writing mistake to take the dimension (3N) of...
I checked again, it shoule be √(N/3) as the original version. The norm calculation should consider 3*N elements, not N elements. Specifically, the norm of the offset should be ||...