teaching-where-to-look icon indicating copy to clipboard operation
teaching-where-to-look copied to clipboard

Published 20 hours ago • verified publisher icon gist-ailab
→

Metadata

Official Implementation of the ECCV 22 paper, "Teaching Where to Look: Attention Similarity Knowledge Distillation for Low Resolution Face Recognition".

Results 6 teaching-where-to-look issues
Sort by recently updated
recently updated
newest added

how to let A-SKD combine with logit distillation (KD)？

can you release the code？thanks your work

Gabriel-18 avatar Gabriel-18

how to draw the attention map,can you release the code

1 comment

ohtiger avatar ohtiger

Hi! Can u offer the trained student model weights trained on 24*24 size image? I don't have enough resource and time to train a model myself.

Same as the title.

saswata-paladhi avatar saswata-paladhi

Can you offer scripts on how to evaluate TinyFace and agedb?

my experiments result is not good .and i want to know why my result is bad

Gabriel-18 avatar Gabriel-18

Why do you train the student model using the last-trained teacher model, rather than the teacher model with the highest recognition rate?"

My reproduced results are significantly different from the results provided in your paper.I evaluated the last student model using the code and parameters provided by you. rank,values 1,42.542918454935624 5,50.88519313304721 10,54.21137339055794...

Gabriel-18 avatar Gabriel-18

Weights links are not working

1 comment

![image](https://github.com/gist-ailab/teaching-where-to-look/assets/29913676/2f7d8df8-df4f-47e3-85cd-2acdd5b33cca)

karanjakhar avatar karanjakhar

← Metadata

53
Stars
8
Forks
Watchers

Owner

verified publisher icon gist-ailab

Metadata

Official Implementation of the ECCV 22 paper, "Teaching Where to Look: Attention Similarity Knowledge Distillation for Low Resolution Face Recognition".

Back