SportsGPT: An LLM-driven Framework for Interpretable Sports Motion Assessment and Training Guidance

2025-12-16

AI Summary

Current intelligent sports analysis systems primarily focus on scoring and visualization, lacking automated performance diagnosis and interpretable training guidance. To address this gap, we propose an end-to-end closed-loop framework for sports motion assessment and coaching recommendation, which takes skeletal time-series data as input and outputs diagnostic conclusions and domain-specific training advice. Our key contributions are threefold: (1) MotionDTW, a novel keyframe alignment algorithm that significantly reduces temporal misalignment and improves Intersection-over-Union (IoU); (2) KISMAM, a knowledge-informed, interpretable assessment model; and (3) SportsRAG, a Retrieval-Augmented Generation framework grounded in a 6B-token sports-domain knowledge base, integrating the Qwen3 large language model with a curated sports knowledge graph. Experiments demonstrate that the synergistic integration of KISMAM and SportsRAG substantially enhances diagnostic accuracy and coaching professionalism, consistently outperforming general-purpose large language models.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Computer Vision: Motion & Tracking
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Existing intelligent sports analysis systems mainly focus on "scoring and visualization," often lacking automatic performance diagnosis and interpretable training guidance. Recent advances of Large Language Models (LMMs) and motion analysis techniques provide new opportunities to address the above limitations. In this paper, we propose SportsGPT, an LLM-driven framework for interpretable sports motion assessment and training guidance, which establishes a closed loop from motion time-series input to professional training guidance. First, given a set of high-quality target models, we introduce MotionDTW, a two-stage time series alignment algorithm designed for accurate keyframe extraction from skeleton-based motion sequences. Subsequently, we design a Knowledge-based Interpretable Sports Motion Assessment Model (KISMAM) to obtain a set of interpretable assessment metrics (e.g., insufficient extension) by constrasting the keyframes with the targe models. Finally, we propose SportsRAG, a RAG-based training guidance model based on Qwen3. Leveraging a 6B-token knowledge base, it prompts the LLM to generate professional training guidance by retrieving domain-specific QA pairs. Experimental results demonstrate that MotionDTW significantly outperforms traditional methods with lower temporal error and higher IoU scores. Furthermore, ablation studies validate the KISMAM and SportsRAG, confirming that SportsGPT surpasses general LLMs in diagnostic accuracy and professionalism.

Problem

• Develops a framework for interpretable sports motion assessment and training guidance
• Extracts keyframes from skeleton-based motion sequences using a two-stage alignment algorithm
• Generates professional training guidance by retrieving domain-specific knowledge from a large database

Innovation

• MotionDTW algorithm extracts keyframes from skeleton motion
• KISMAM model provides interpretable metrics by comparing keyframes
• SportsRAG uses RAG with Qwen3 to generate training guidance