HumMUSS: Human Motion Understanding using State Space Models

arXiv:2404.10880v1 Announce Type: new
Abstract: Understanding human motion from video is essential for a range of applications, including pose estimation, mesh recovery and action recognition. While state-of-the-art methods predominantly rely on transformer-based architectures, these approaches have limitations in practical scenarios. Transformers are slower when sequentially predicting on a continuous stream of frames in real-time, and do not generalize to new frame rates. In light of these constraints, we propose a novel attention-free spatiotemporal model for human motion understanding building upon …

abstract action recognition applications architectures art arxiv continuous cs.ai cs.cv human limitations mesh practical recognition recovery space state state space models transformer transformers type understanding video