We present a semantics modulated, multi scale Transformer for 3D gaze estimation. Our model conditions CLIP global features with learnable prototype banks (illumination, head pose, background, direction), fuses these prototype-enriched global vectors with CLIP patch tokens and high-resolution CNN tokens in a unified attention space, and replaces several FFN blocks with routed/shared Mixture of Experts to increase conditional capacity. Evaluated on MPIIFaceGaze, EYEDIAP, Gaze360 and ETH-XGaze, our model achieves new state of the art angular errors of 2.49{\deg}, 3.22{\deg}, 10.16{\deg}, and 1.44{\deg}, demonstrating up to a 64% relative improvement over previously reported results. ablations attribute gains to prototype conditioning, cross scale fusion, MoE and hyperparameter. Our code is publicly available at https://github. com/AIPMLab/Gazeformer.
@article{arxiv.2601.12316,
title = {GazeFormer-MoE: Context-Aware Gaze Estimation via CLIP and MoE Transformer},
author = {Xinyuan Zhao and Xianrui Chen and Ahmad Chaddad},
journal= {arXiv preprint arXiv:2601.12316},
year = {2026}
}