Does the img_fake_att of decoder represent x_^ in the paper? - Githubissues

ZhaoJ9014 / High-Performance-Face-Recognition

🔥🔥Several High-Performance Models for Unconstrained/Large-Scale/Low-Shot Face Recognition🔥🔥

MIT License

361 stars 79 forks source link

Does the img_fake_att of decoder represent x_^ in the paper? #11

Open chengxiqiao opened 5 years ago

chengxiqiao commented 5 years ago

If it does, the code "return x, img att + x (1. - att), att" is contrary to the Eq.(13) in paper

chengxiqiao commented 5 years ago

Is the code complete? I'm curious about how to keep the identity information @ZhaoJ9014

Hzzone commented 4 years ago

img_fake_att is the attention mask which is first proposed in [GANimation: Anatomically-aware Facial Animation from a Single Image, eccv 2018]. I am confused about why the code in https://github.com/ZhaoJ9014/High-Performance-Face-Recognition/blob/a0b6b6b269092051afb7a09a0100409cbccfa493/src/Look%20Across%20Elapse-%20Disentangled%20Representation%20Learning%20and%20Photorealistic%20Cross-Age%20Face%20Synthesis%20for%20Age-Invariant%20Face%20Recognition.TensorFlow/network.py#L150 has fed it into D_img which makes no sense....

Hzzone commented 4 years ago

Besides, I have not found $L_{ip}$ in this code.....😥😥😥😥