HyLiFormer: Hyperbolic Linear Attention for Skeleton-based Human Action Recognition

Y Li, H Qu, M Liu, J Liu, Y Cai - arxiv preprint arxiv:2502.05869, 2025‏ - arxiv.org
Transformers have demonstrated remarkable performance in skeleton-based human action
recognition, yet their quadratic computational complexity remains a bottleneck for real-world …

SkelMamba: A State Space Model for Efficient Skeleton Action Recognition of Neurological Disorders

N Martinel, M Serrao, C Micheloni - arxiv preprint arxiv:2411.19544, 2024‏ - arxiv.org
We introduce a novel state-space model (SSM)-based framework for skeleton-based human
action recognition, with an anatomically-guided architecture that improves state-of-the-art …

FreqMixFormerV2: Lightweight Frequency-aware Mixed Transformer for Human Skeleton Action Recognition

W Wu, P Wang, C Chen, A Lu - arxiv preprint arxiv:2412.20621, 2024‏ - arxiv.org
Transformer-based human skeleton action recognition has been developed for years.
However, the complexity and high parameter count demands of these models hinder their …