Learn From All: Erasing Attention Consistency for Noisy Label Facial Expression Recognition

21 Jul 2022  ·  Yuhang Zhang, Chengrui Wang, Xu Ling, Weihong Deng ·

Noisy label Facial Expression Recognition (FER) is more challenging than traditional noisy label classification tasks due to the inter-class similarity and the annotation ambiguity. Recent works mainly tackle this problem by filtering out large-loss samples. In this paper, we explore dealing with noisy labels from a new feature-learning perspective. We find that FER models remember noisy samples by focusing on a part of the features that can be considered related to the noisy labels instead of learning from the whole features that lead to the latent truth. Inspired by that, we propose a novel Erasing Attention Consistency (EAC) method to suppress the noisy samples during the training process automatically. Specifically, we first utilize the flip semantic consistency of facial images to design an imbalanced framework. We then randomly erase input images and use flip attention consistency to prevent the model from focusing on a part of the features. EAC significantly outperforms state-of-the-art noisy label FER methods and generalizes well to other tasks with a large number of classes like CIFAR100 and Tiny-ImageNet. The code is available at https://github.com/zyh-uaiaaaa/Erasing-Attention-Consistency.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Facial Expression Recognition (FER) Acted Facial Expressions In The Wild (AFEW) EAC Accuracy(on validation set) 65.32% # 3
Facial Expression Recognition (FER) AffectNet EAC Accuracy (7 emotion) 65.32 # 13
Facial Expression Recognition (FER) FER+ EAC Accuracy 89.64 # 7
Facial Expression Recognition (FER) RAF-DB EAC(ResNet-50) Overall Accuracy 90.35 # 8

Methods