On a new NUGGET gesture dataset, skeleton-based HD-GCN beats vision foundation model V-JEPA (94.4% vs 90.1% top-1), while zero-shot Gemini Flash 2.0 achieves only 42.1%.
HGR-ViT: Hand gesture recognition with vision transformer,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
How do Foundation Models Compare to Skeleton-Based Approaches for Gesture Recognition in Human-Robot Interaction?
On a new NUGGET gesture dataset, skeleton-based HD-GCN beats vision foundation model V-JEPA (94.4% vs 90.1% top-1), while zero-shot Gemini Flash 2.0 achieves only 42.1%.