CrossGLG: LLM Guides One-shot Skeleton-based 3D Action Recognition in a Cross-level Manner

arXiv:2403.10082v1 Announce Type: new
Abstract: Most existing one-shot skeleton-based action recognition focuses on raw low-level information (e.g., joint location), and may suffer from local information loss and low generalization ability. To alleviate these, we propose to leverage text description generated from large language models (LLM) that contain high-level human knowledge, to guide feature learning, in a global-local-global way. Particularly, during training, we design $2$ prompts to gain global and local text descriptions of each action from an LLM. We first …

abstract action recognition arxiv cs.cv generated guides information language language models large language large language models llm location loss low raw recognition text type