VGNet training generate the some faces of each identity

[Erdos001]

First very thanks for release the code I use the project to train the vgnet and encounter the issue that Gen Loss keeps increasing and the vgnet model generates the same faces of as the I_p of each identity, could you suggest what would be the reason?

[lelechen63]

First very thanks for release the code I use the project to train the vgnet and encounter the issue that Gen Loss keeps increasing and the vgnet model generates the same faces of as the I_p of each identity, could you suggest what would be the reason?

Hi there, I will look into this after eccv deadline. If you need to use it for testing, please use the pre-trained model, which is a universal model that can be tested on any video.