Gaze Representation

Gaze representation research focuses on computationally modeling and accurately estimating eye gaze direction from visual data, primarily aiming to improve the robustness and efficiency of gaze estimation systems. Current research heavily emphasizes unsupervised and self-supervised learning techniques, employing contrastive learning and innovative architectures like capsule networks and multi-stream networks to overcome the limitations of data scarcity and computational cost associated with supervised approaches. These advancements have significant implications for various fields, including human-computer interaction, assistive technologies, and emotion recognition, by enabling more accurate and efficient analysis of human attention and behavior.