Artificial Intelligence Applications in Table Tennis Biomechanics: A Systematic Review from a Sports Engineering Perspective

초록

This systematic review synthesized recent evidence on the application of artificial intelligence (AI) in table tennis biomechanics from a sports engineering perspective. The review was conducted in accordance with the PRISMA 2020 framework. Four major databases, including Web of Science, PubMed, ScienceDirect, and Google Scholar, were searched for studies published between January 2016 and September 2025. After screening and eligibility assessment, 22 studies were included in the final review. The included studies were grouped into three main domains: (1) motion recognition and classification, (2) technical assessment and sports performance analysis, and (3) intelligent training assistance and real-time feedback. Across these domains, AI applications primarily relied on computer vision, wearable inertial sensors, pose estimation methods, and multimodal data processing frameworks. The reviewed evidence suggests that AI has improved the automation and objectivity of movement analysis and has expanded the potential for more data-driven training support in table tennis. However, the current evidence base remains methodologically heterogeneous and is constrained by small sample sizes, reliance on private datasets, inconsistent validation procedures, and limited real-world testing. From a sports engineering perspective, the field is moving toward more integrated and practically deployable systems that combine sensing, computation, and feedback in training environments. Future research should prioritize multimodal data fusion, stronger external validation, dataset standardization, interpretable AI frameworks, and lightweight real-time deployment to improve reproducibility, robustness, and practical applicability.

키워드