search

zyh-uaiaaaa / Erasing-Attention-Consistency

Official implementation of the ECCV2022 paper: Learn From All: Erasing Attention Consistency for Noisy Label Facial Expression Recognition
77 stars 15 forks source link

Pre-trained model? #2

Closed chi0tzp closed 1 year ago

chi0tzp commented 1 year ago

Hi, thanks for making the code public! Any thoughts on providing a pre-trained code and a testing script soon? Thanks!

zyh-uaiaaaa commented 1 year ago

Hi, the pre-trained model is borrowed from https://github.com/ox-vgg/vgg_face2/blob/master/README.md, which is pre-trained on MS-Celeb-1M [1]. As we mainly deal with noisy label learning of facial expression recognition, we follow [2] and test after each training epoch, the accuracy is reported as the mean accuracy of the last 5 epochs.

[1] Y. Guo, L. Zhang, Y. Hu, X. He, and J. Gao. MS-Celeb-1M: A Dataset and Benchmark for Large Scale Face Recognition. In ECCV 2016. [2] Wang, K., Peng, X., Yang, J., Lu, S., Qiao, Y.: Suppressing uncertainties for large-scale facial expression recognition. In CVPR 2020.