AI RESEARCH
GMGaze: MoE-Based Context-Aware Gaze Estimation with CLIP and Multiscale Transformer
arXiv CS.CV
•
ArXi:2605.00799v1 Announce Type: new Gaze estimation methods commonly use facial appearances to predict the direction of a person gaze. However, previous studies show three major challenges with convolutional neural network (CNN)-based, transformer-based, and contrastive language-image pre-