HaGRIDv2: 1M Images for Static and Dynamic Hand Gesture Recognition
This work provides an incremental enhancement to hand gesture recognition datasets, benefiting researchers and developers in human-computer interaction and related fields.
The paper introduces HaGRIDv2, an expanded dataset with 1M images for static and dynamic hand gesture recognition, adding 15 new gestures and improving the 'no gesture' class to reduce false positives by 6 times, achieving the best generalization ability among gesture datasets.
This paper proposes the second version of the widespread Hand Gesture Recognition dataset HaGRID -- HaGRIDv2. We cover 15 new gestures with conversation and control functions, including two-handed ones. Building on the foundational concepts proposed by HaGRID's authors, we implemented the dynamic gesture recognition algorithm and further enhanced it by adding three new groups of manipulation gestures. The ``no gesture" class was diversified by adding samples of natural hand movements, which allowed us to minimize false positives by 6 times. Combining extra samples with HaGRID, the received version outperforms the original in pre-training models for gesture-related tasks. Besides, we achieved the best generalization ability among gesture and hand detection datasets. In addition, the second version enhances the quality of the gestures generated by the diffusion model. HaGRIDv2, pre-trained models, and a dynamic gesture recognition algorithm are publicly available.