zhaoj9014 / high-performance-face-recognition Goto Github PK

In your code /corss-age/network.py -> line:156 function lossFunc()
You write that 0.00*tvLoss(img_fake), I wonder if it is 0.001 ？？

在您的的人脸老化代码中，network.py156行0.00*tvLoss(img_fake)的0.00是写错了么，我觉得应该是0.001吧，不然没有必要写

Here is the quick view:https://github.com/ZhaoJ9014/High-Performance-Face-Recognition/blob/master/src/Look%20Across%20Elapse-%20Disentangled%20Representation%20Learning%20and%20Photorealistic%20Cross-Age%20Face%20Synthesis%20for%20Age-Invariant%20Face%20Recognition.TensorFlow/network.py#L156

thanks for your research. when i read your paper ,it says that the dataset will release soon. where can i find the dataset? hope your replay.

Which model did you use to achieve 99.85% on LFW? Is it LightCNN?

http://47.95.194.180/

Hi @ZhaoJ9014 could you please share the demo/inference source code of AIM? thank you.

this one:
Look Across Elapse- Disentangled Representation Learning and Photorealistic Cross-Age Face Synthesis for Age-Invariant Face Recognition.TensorFlow.
it seems that there is only training code.

Hello Dears,

I'm unable to test this model, I was just able to train it. If someone did test it, please share testing steps

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

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!

may you offer the environment to run your code? such as the version of tensorflow.
and may you tell when you will release the dataset?
thank you very much! @ZhaoJ9014

Invalid argument: Assign requires shapes of both tensors to match. lhs shape= [250,16384] rhs shape= [150,16384]
[[Node: save/Assign_68 = Assign[T=DT_FLOAT, _class=["loc:@G_fc/w"], use_locking=true, validate_shape=true, _device="/job:localhost/replica:0/task:0/gpu:0"]

Could you please show the test code of AIM？There are only train code about train code.

Is there a plan to release pretrained model for look across elapse? The training seems very slow with the given script. My device: RTX2080 Ti.

Hi, Zhao
Thanks for your amazing work! And will the CAFR dataset be released soon?

Thanks!

Is there a PyTorch implementation for the model?

Can you publish data sets and models in Baidu cloud? Google hard disk downloads are all wrong, prompting too many people to download recently, has been blocked. Thank you

Download
Please refer to "./src/" and this link to access our full source codes and models (continue to update).
https://drive.google.com/drive/folders/15L-macCtlFDondqbDDoZFcz1Qe9bLqH4?usp=sharing

How can I test the model? I see that all the other modules on your github have their own training and testing part.