showlab / byoc Goto Github PK

感谢作者开源代码。
请问，landmark2exp/checkpoints/facerecon/results/test_images/epoch_20_000000文件需要怎么获取呢？

Nice work! This work is very facinating, however, I got some problems while trying to reconstruct a talking head video.

Since I cannot find any video results related to this work, I tried to reconstruct a video by myself. However, the facial movements are quite strange in the recon video. This is what I got, in which the lips seems to be sticked:

I wonder is there anything wrong in my pipeline:

1. I used image2bs/inference.py to generate predicted_blendshape.csv
   predicted_blendshape.csv

2. For those 50-dimension blendshapes, I directly changed the numbers on the right from shape_1_channel to shape_50_channel corresponding to the keys in the csv file from left to right.
   

Thank you for any possible help!

Hi,
I just found the size of output blendshaeps is 50 in file predicted_blendshape.csv, which is not consistant with arkit, how can I correspond the blendshape1 in character M111.fbx to standard arkit blendshapes ?

Since there are many different versions of blendshapes, and ARKit is one of them, I would like to know if the 50-dimensional blendshape coefficients generated in your work can correspond to ARKit, similar to #3
If so, the generated blendshape coefficients can be used on different 3D models.
If not, I would also like to know which format your blendshape coefficients conform to, and how can I find any 3D models online that have the same blendshape keys as the male FBX file you provided?
Thank you very much.