GMGaze achieves mean angular errors of 2.49°, 3.22°, 10.16°, and 1.44° on MPIIFaceGaze, EYEDIAP, Gaze360, and ETH-XGaze by early context-conditioned fusion and MoE scaling, outperforming baselines in within- and cross-domain settings.
Learning to prompt for vision-language models.International journal of computer vision, 130(9):2337–2348, 2022
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
GMGaze: MoE-Based Context-Aware Gaze Estimation with CLIP and Multiscale Transformer
GMGaze achieves mean angular errors of 2.49°, 3.22°, 10.16°, and 1.44° on MPIIFaceGaze, EYEDIAP, Gaze360, and ETH-XGaze by early context-conditioned fusion and MoE scaling, outperforming baselines in within- and cross-domain settings.