Facial Expression Recognition Using Residual Masking Network

2021-01-10

International Conference on Pattern Recognition

AI Summary

This work addresses the limited accuracy and robustness of facial expression recognition in complex scenarios by proposing a residual mask network architecture that integrates an explicit attention mechanism into convolutional neural networks through segmentation masks. The approach combines deep residual networks with a U-Net-like structure to generate refined feature masks corresponding to expression-relevant facial regions, thereby guiding the model to focus on discriminative areas and enhancing feature learning. Experimental results on the FER2013 benchmark and a private VEMO dataset demonstrate that the proposed method achieves state-of-the-art recognition accuracy, significantly improving model performance under challenging environmental conditions.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Deep Neural Architectures and Foundation Models
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Responsible Web: Algorithmic accountability and transparency on the web
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Automatic facial expression recognition (FER) has gained much attention due to its applications in human-computer interaction. Among the approaches to improve FER tasks, this paper focuses on deep architecture with the attention mechanism. We propose a novel Masking Idea to boost the performance of CNN in facial expression task. It uses a segmentation network to refine feature maps, enabling the network to focus on relevant information to make correct decisions. In experiments, we combine the ubiquitous Deep Residual Network and Unet-like architecture to produce a Residual Masking Network. The proposed method holds state-of-the-art (SOTA) accuracy on the well-known FER2013 and private VEMO datasets.

Problem

• Facial Expression Recognition
• Human-Computer Interaction
• Deep Learning
• Attention Mechanism

Innovation

• Residual Masking Network
• Facial Expression Recognition
• Attention Mechanism
• Feature Refinement
• Deep Residual Network