zqbai-jeremy

> Really interesting work! > > There is one question that came up to me about this general approach of modelling a scene with an MLP compared to a general...

Did you change any code? Look like "expPC.fromfile(Expbin, 3 * exp_dim[0] * n_vertex)" is in line 57. BTW, I did not remember I faced this error before.

Sorry for my late reply. You can try this repo (https://github.com/zqbai-jeremy/face3d), which has minor modifications comparing to the original one from YadiraF/face3d.

Sorry there seems a bug I forgot to fix. They should be model path. The former one is for building rig and the second is for video tracking.

> checkpoint_path and video_checkpoint_path in demo.py are confusing Fixed. Should work now.

Thank you for your interest in our work! Sorry that current model cannot be rigged in Apple 52 blendshape.

Thank you very much for your interest in our work! You can find the steps of setting up BFM here: https://github.com/zqbai-jeremy/face3d/tree/master/examples/Data/BFM

Need to run the matlab script. See step 4.

我们用的Stirling/ESRC 3D face database (http://pics.stir.ac.uk/ESRC/)进行的训练，paper里section 4.1有大致的训练数据生成过程。

抱歉回复晚了。这个我不太确定，但如果是styleGAN那种级别的fake图，我感觉可能没有很明显的差异。