Convolutional Network with Mamba State Space Models for Skeleton-based Action Recognition
This work proposes a novel HAR framework that integrates GCNs with an Adaptive Adjacency Matrix for spatial modeling and the Mamba State Space Model for temporal feature extraction, which achieves competitive performance with an approximately 72% reduction in parameter count.
Amine Mansouri, Abdellah Elzaar, Toufik Bakir et al.
· 0 citations