SCALE: Semantic- and Confidence-Aware Conditional Variational Autoencoder for Zero-shot Skeleton-based Action Recognition

SCALE: Semantic- and Confidence-Aware Conditional Variational Autoencoder for Zero-shot Skeleton-based Action Recognition explores A novel framework for zero-shot skeleton-based action recognition that leverages conditional variational autoencoders and a confidence-aware energy loss to improve accuracy without explicit skeleton-text alignment.. Commercial viability score: 5/10 in Computer Vision.