Posture-Driven Action Intent Inference for Playing style and Fatigue Assessment

2025-07-15

AI Summary

This study addresses the challenge of inferring athletes’ action intentions from video data. We propose a pose-driven weakly supervised learning framework for racket-sport intent recognition—specifically, distinguishing offensive versus defensive stroke styles—and fatigue state assessment. To mitigate difficulties in annotating sensitive human motion videos and comply with privacy constraints, our method jointly leverages single-frame pose estimation and temporal modeling of pose sequences, exploiting naturally diverse, emotion-conditioned pose trajectories in sports scenarios—without requiring fine-grained action labels. Experiments on real-world match videos—characterized by high noise and variability—demonstrate strong robustness: the model achieves an F1 score of 75.3% and an AUC-ROC of 82.1%, significantly outperforming fully supervised baselines. To our knowledge, this is the first work to systematically apply weakly supervised temporal pose modeling to competitive intent understanding, establishing a novel paradigm for non-intrusive, interpretable sports behavior analysis.

Technology Category

• Computer Vision: Video Understanding & Activity Analysis
• Intelligent Robots: State Estimation
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Posture-based mental state inference has significant potential in diagnosing fatigue, preventing injury, and enhancing performance across various domains. Such tools must be research-validated with large datasets before being translated into practice. Unfortunately, such vision diagnosis faces serious challenges due to the sensitivity of human subject data. To address this, we identify sports settings as a viable alternative for accumulating data from human subjects experiencing diverse emotional states. We test our hypothesis in the game of cricket and present a posture-based solution to identify human intent from activity videos. Our method achieves over 75% F1 score and over 80% AUC-ROC in discriminating aggressive and defensive shot intent through motion analysis. These findings indicate that posture leaks out strong signals for intent inference, even with inherent noise in the data pipeline. Furthermore, we utilize existing data statistics as weak supervision to validate our findings, offering a potential solution for overcoming data labelling limitations. This research contributes to generalizable techniques for sports analytics and also opens possibilities for applying human behavior analysis across various fields.

Problem

• Inferring action intent from posture for style and fatigue assessment
• Validating posture-based mental state inference with large datasets
• Overcoming data sensitivity challenges in vision-based diagnosis

Innovation

• Posture-based intent inference from activity videos
• Motion analysis for aggressive and defensive shot identification
• Weak supervision using existing data statistics