search

ZhaoJ9014 / High-Performance-Face-Recognition

🔥🔥Several High-Performance Models for Unconstrained/Large-Scale/Low-Shot Face Recognition🔥🔥
MIT License
360 stars 79 forks source link

CAFR dataset release #3

Open crystal0516 opened 5 years ago

crystal0516 commented 5 years ago

Hello, Dr. Zhao. Thank you for awesome work and sharing AIM codes. I have trained and tested on some public (e.g. FGNet) database with your codes. However, as desribed in your paper, training on the CAFR dataset is very important for best recognition performance. Will the CAFR database be released? Thank you!

dyk-ml commented 5 years ago

How to test on other databases for face recognition? i just generated some pictures in different ages through this code.can you share some points?

chengxiqiao commented 5 years ago

Which version do you use when traning the codes?

ccl-private commented 5 years ago

@dyk-ml In my opinion, self.encoder(x) can extract a face coder for face recognition. we only need to compare this vector to recognize a person. But I think the trained result of mine was not that good. I don't know do you have the same problem. This is one example: Screenshot from 2019-09-11 16-58-38 0_0.jpg to 0_9.jpg is not that like 0_r.jpg, they are more like an average of the whole data set.

ReverseSystem001 commented 4 years ago

Hello, Dr. Zhao. Thank you for awesome work and sharing AIM codes. I have trained and tested on some public (e.g. FGNet) database with your codes. However, as desribed in your paper, training on the CAFR dataset is very important for best recognition performance. Will the CAFR database be released? Thank you!

SO, Does the CAFR database released?

ReverseSystem001 commented 4 years ago

@dyk-ml In my opinion, self.encoder(x) can extract a face coder for face recognition. we only need to compare this vector to recognize a person. But I think the trained result of mine was not that good. I don't know do you have the same problem. This is one example: Screenshot from 2019-09-11 16-58-38 0_0.jpg to 0_9.jpg is not that like 0_r.jpg, they are more like an average of the whole data set.

have you been solved the problem??? hope your replay.

EddyRonald commented 2 years ago

@crystal0516 I am trying to run AIM with the FGNET dataset but I have a problem like this: ValueError: not enough values to unpack (expected 3, got 0) Could you tell me how I get to compile the code with FGNET please.