In the field of 3-D action recognition, self-supervised learning has shown promising results but remains a challenging task. Previous approaches to motion modeling often relied on selecting features ...

This paper proposes a novel method for human skeleton-based action recognition using a Topology Unshared Graph Convolutional Network (TU-GCN). This approach enhances recognition accuracy and network ...

BitSummit 2025 reveals nominees across eight award categories! Discover standout indie titles, demo them in Kyoto & cast your ...