What is exp_ori in the given dataset?
ChikaYan opened this issue · 2 comments
Hi,
Thank you so much for the excellent code and for at least making the inference code avaliable!
May I ask a quick question -- in the released dataset there seems to be a exp_ori, which is passed all the way to the forward(), but is actually not used (only exp is used). May I ask what exactly is exp_ori?
Many thanks!
Hi, thanks for your interest in our work.
Only exp_ori is used in our implementation. You could see the following reference to verify this:
https://github.com/USTC3DV/NeRFBlendShape-code/blob/d198ac270b9994e36d7ef86003777a2c1976ca7a/nerf/provider.py#L131C46-L131C53
exp_ori is 46-dim, which represents the extent of blendshape's 46 AU activations. But in some video, not every AU could activated, so exp is a subset of exp_ori that could be actually activated in one video.
Selecting a subset is not necessary, so you could ignore exp because it is not used in this code anymore.
Ah I see, thanks a lot for the detailed explaination!