TL;DR
A novel contrastive learning framework, PIA-ACL, which explicitly model the person dimension to construct domain-level local interaction representations, enhancing the model’s capability in understanding multi-person actions and proposing an Asymmetric Multi-Query Augmentation strategy to facilitate contrastive learning.
Skeleton-based action recognition has gained increasing attention as a robust alternative to RGB-based methods due to its resilience to background clutter and lighting variations. Existing approaches have achieved notable progress, particularly with graph-based models that capture topological and temporal patterns of h...
Multimodal human activity recognition (HAR) benefits from complementary skeleton, inertial, and visual observations. However, many learning-based models still treat relationships among modalities, joints, and sensor variables as symmetric associations. This limits their ability to represent asymmetric information flow...
A novel Mask-aware Action Spatiotemporal Quantization framework that decouples the conflicting tasks of spatial feature inference and temporal smoothing, and establishes a comprehensive and substantial leading advantage in the Mean over Frames accuracy.
Weakly-supervised temporal action localization aims to localize action instances in untrimmed videos using only video-level annotations. However, due to the limited representational capacity of single-modal information and the absence of temporal annotations, the localization framework struggles to extract representati...
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
