N-UCLA

The N-UCLA dataset is a benchmark for skeleton-based action recognition that is used to evaluate continual graph learning methods in the context of spatio-temporal graphs.

Papers using N-UCLA (3)

Adaptive Hyper-Graph Convolution Network for Skeleton-based Human Action Recognition with Virtual Connections2024 · 12 citesTopological Symmetry Enhanced Graph Convolution for Skeleton-Based Action Recognition2024Benchmarking Sensitivity of Continual Graph Learning for Skeleton-Based Action Recognition2024