B-MoE framework achieves state-of-the-art performance on micro-action recognition by using region-specific experts and cross-attention routing.
Quo vadis, action recognition? a new model and the kinetics dataset
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.CV 3verdicts
UNVERDICTED 3representative citing papers
A gaze-only student model distilled from a joint gaze-video teacher achieves high skill-assessment accuracy using 73x less power than prior methods.
A new dataset with high-fidelity close-up garment images and full/close-up try-on videos plus the VGID metric enables better texture and structure preservation in high-resolution video virtual try-on.
citing papers explorer
-
B-MoE: A Body-Part-Aware Mixture-of-Experts "All Parts Matter" Approach to Micro-Action Recognition
B-MoE framework achieves state-of-the-art performance on micro-action recognition by using region-specific experts and cross-attention routing.
-
SkillSight: Efficient First-Person Skill Assessment with Gaze
A gaze-only student model distilled from a joint gaze-video teacher achieves high skill-assessment accuracy using 73x less power than prior methods.
-
Eevee: Towards Close-up High-resolution Video-based Virtual Try-on
A new dataset with high-fidelity close-up garment images and full/close-up try-on videos plus the VGID metric enables better texture and structure preservation in high-resolution video virtual try-on.