Exploring Large-scale Unlabeled Faces to Enhance Facial Expression Recognition
This work addresses dataset limitations for FER applications in human-computer interaction and emotion analysis, but it is incremental as it builds on semi-supervised methods.
The paper tackled the problem of limited dataset size in Facial Expression Recognition (FER) by proposing a semi-supervised learning framework that uses unlabeled face data, achieving excellent results on the ABAW5 EXPR validation set.
Facial Expression Recognition (FER) is an important task in computer vision and has wide applications in human-computer interaction, intelligent security, emotion analysis, and other fields. However, the limited size of FER datasets limits the generalization ability of expression recognition models, resulting in ineffective model performance. To address this problem, we propose a semi-supervised learning framework that utilizes unlabeled face data to train expression recognition models effectively. Our method uses a dynamic threshold module (\textbf{DTM}) that can adaptively adjust the confidence threshold to fully utilize the face recognition (FR) data to generate pseudo-labels, thus improving the model's ability to model facial expressions. In the ABAW5 EXPR task, our method achieved excellent results on the official validation set.