LLMs are Good Action Recognizers

arXiv:2404.00532v1 Announce Type: new
Abstract: Skeleton-based action recognition has attracted lots of research attention. Recently, to build an accurate skeleton-based action recognizer, a variety of works have been proposed. Among them, some works use large model architectures as backbones of their recognizers to boost the skeleton data representation capability, while some other works pre-train their recognizers on external data to enrich the knowledge. In this work, we observe that large language models which have been extensively used in various natural …

abstract action recognition architectures arxiv attention boost build capability cs.cv data good llms recognition representation research them train type