Emotion Recognition and Modeling Advances

The field of emotion recognition and modeling is moving towards more nuanced and personalized approaches. Researchers are exploring new methods to capture the complexities of human emotions, including the use of geometric frameworks, graph-based modeling, and topological approaches. The development of novel datasets and benchmarks is also facilitating progress in this area, enabling the evaluation of models in more realistic and diverse settings. Notably, some studies are investigating the relationship between emotions and other factors, such as memorability and perceptual variability, to improve the accuracy of emotion recognition systems. Noteworthy papers include: The paper on the Coordinate Heart System, which introduces a geometric framework for emotion representation and provides a comprehensive computational framework for AI emotion recognition. The paper on Synthesizing Images on Perceptual Boundaries of ANNs, which investigates the phenomenon of high perceptual variability in human perception and proposes a novel perceptual boundary sampling method to generate facial expression stimuli.

Sources

The Emotion-Memory Link: Do Memorability Annotations Matter for Intelligent Systems?

Synthesizing Images on Perceptual Boundaries of ANNs for Uncovering Human Perceptual Variability on Facial Expressions

Coordinate Heart System: A Geometric Framework for Emotion Representation

Exp-Graph: How Connections Learn Facial Attributes in Graph-based Expression Recognition

Hybrid-supervised Hypergraph-enhanced Transformer for Micro-gesture Based Emotion Recognition

Rethinking Occlusion in FER: A Semantic-Aware Perspective and Go Beyond

Salience Adjustment for Context-Based Emotion Recognition

Persistent Patterns in Eye Movements: A Topological Approach to Emotion Recognition

OPEN: A Benchmark Dataset and Baseline for Older Adult Patient Engagement Recognition in Virtual Rehabilitation Learning Environments

Built with on top of