Wenzheng Zeng

Hi! I have the same question about the correspondence between the 38365 indexs and its actual position on face. Anyone could help?

Thanks for your great work. Can you release the code of the near online inference? A demo code with a smaller input clip length (e.g., 15) on YouTubeVIS2019 would be...

> About the seconde question, I have found the answer. Torch.nn.CrossEntropyLoss() combines nn.LogSoftmax() and nn.NLLLoss() in one single class. So is the implementation here of ArcFace loss right?

Thanks for your great work! Can you provide a demo for face alignment?

@jiangxiangchuan I obtain a similar result on 3090. I think the reported time (i.e., 4.9ms) just contains the session.run function. The time for data pre-processing is not included.

Can this code work with torch>1.7.0? The reason is that my CUDA version is 11, which may not work with torch 1.4. I have tried to run setup.py but failed,...