ChatMotion: A Multimodal Multi-Agent for Human Motion Analysis
This work addresses the problem of interactivity and adaptability in human motion analysis for users in fields like healthcare or robotics, representing an incremental advancement over existing MLLMs.
The paper tackles the limitations of 'instruct-only' multimodal large language models in human motion analysis by introducing ChatMotion, a multimodal multi-agent framework that dynamically interprets user intent and decomposes tasks, resulting in demonstrated improvements in precision, adaptability, and user engagement.
Advancements in Multimodal Large Language Models (MLLMs) have improved human motion understanding. However, these models remain constrained by their "instruct-only" nature, lacking interactivity and adaptability for diverse analytical perspectives. To address these challenges, we introduce ChatMotion, a multimodal multi-agent framework for human motion analysis. ChatMotion dynamically interprets user intent, decomposes complex tasks into meta-tasks, and activates specialized function modules for motion comprehension. It integrates multiple specialized modules, such as the MotionCore, to analyze human motion from various perspectives. Extensive experiments demonstrate ChatMotion's precision, adaptability, and user engagement for human motion understanding.