Privileged Attribution Constrained Deep Networks for Facial Expression Recognition
This addresses the challenge of noisy data and small datasets in facial expression recognition, which is crucial for machines to understand human behaviors, but it appears incremental as it builds on existing attention mechanisms.
The paper tackles the problem of facial expression recognition by proposing a method to guide models to focus on specific facial areas like eyes and mouth, and it reports outperforming current state-of-the-art methods on RAF-DB and AffectNet datasets.
Facial Expression Recognition (FER) is crucial in many research domains because it enables machines to better understand human behaviours. FER methods face the problems of relatively small datasets and noisy data that don't allow classical networks to generalize well. To alleviate these issues, we guide the model to concentrate on specific facial areas like the eyes, the mouth or the eyebrows, which we argue are decisive to recognise facial expressions. We propose the Privileged Attribution Loss (PAL), a method that directs the attention of the model towards the most salient facial regions by encouraging its attribution maps to correspond to a heatmap formed by facial landmarks. Furthermore, we introduce several channel strategies that allow the model to have more degrees of freedom. The proposed method is independent of the backbone architecture and doesn't need additional semantic information at test time. Finally, experimental results show that the proposed PAL method outperforms current state-of-the-art methods on both RAF-DB and AffectNet.