{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,8]],"date-time":"2026-10-08T18:05:22Z","timestamp":1791482722340,"version":"4.3.4"},"reference-count":26,"publisher":"Public Library of Science (PLoS)","issue":"5","license":[{"start":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T00:00:00Z","timestamp":1746662400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Centre of Excellence for Sports Science and Analytics for funding from the Indian Institute of Technology, Madras","award":["SP22231231CPETWOSSAHOC"],"award-info":[{"award-number":["SP22231231CPETWOSSAHOC"]}]}],"content-domain":{"domain":["www.plosone.org"],"crossmark-restriction":false},"short-container-title":["PLoS One"],"abstract":"<jats:p>Boxing punch type classification and kinematic analysis are essential for coaches and athletes, providing critical insights into punch variety and effectiveness, which are vital for performance improvement. Existing methods for punch recognition and classification typically rely on wearable sensor data or video data; however, no fully automated system currently exists. While coaches prefer video-based analysis for its ability to easily visualize punch action errors and refine technique, video-based classification suffers from lower accuracy compared to sensor-based methods due to limitations such as motion blur. Current classification approaches typically employ supervised learning, requiring experts to annotate 70\u201380% of the data for model training. However, the high sampling frequency of sensor data makes this process time-consuming and challenging, leading to potential fatigue and an increased risk of inconsistent annotations by domain experts. This paper proposes a novel multimodal approach that integrates wearable sensor data and video data for automatic punch recognition and classification. The method also includes automatic segmentation of punch videos, which improves classification accuracy by utilizing both data sources. To reduce labeling effort, we apply a Query by Committee-based active learning technique, significantly decreasing the required labeling effort by one-sixth. Using only 15% of the typical labeling effort, our system achieves 91.41% accuracy for rear-hand punch recognition, 91.91% for lead-hand punch recognition, and 92.33% and 94.56% for punch classification, respectively. This Smart Boxer system aims to enhance punch analytics in boxing, providing valuable insights to improve training, optimize performance, and increase fan engagement with the sport.<\/jats:p>","DOI":"10.1371\/journal.pone.0322490","type":"journal-article","created":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T17:36:54Z","timestamp":1746725814000},"page":"e0322490","update-policy":"https:\/\/doi.org\/10.1371\/journal.pone.corrections_policy","source":"Crossref","is-referenced-by-count":6,"title":["An active machine learning framework for automatic boxing punch recognition and classification using upper limb kinematics"],"prefix":"10.1371","volume":"20","author":[{"given":"Saravanan","family":"Manoharan","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"John","family":"Warburton","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ravi Sadananda","family":"Hegde","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ranganathan","family":"Srinivasan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9048-8692","authenticated-orcid":true,"given":"Babji","family":"Srinivasan","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2025,5,8]]},"reference":[{"issue":"6","key":"pone.0322490.ref001","doi-asserted-by":"crossref","DOI":"10.1016\/j.heliyon.2022.e09633","article-title":"An overview of Human Action Recognition in sports based on Computer Vision","volume":"8","author":"K Host","year":"2022","journal-title":"Heliyon"},{"key":"pone.0322490.ref002","first-page":"3","article-title":"Non-invasive performance measurement in combat sports","author":"S Behendi","year":"2016","journal-title":"Proceedings of the 10th International Symposium on Computer Science in Sports (ISCSS)"},{"key":"pone.0322490.ref003","first-page":"4545","article-title":"Combat sports analytics: Boxing punch classification using overhead depthimagery","author":"S Kasiri-Bidhendi","year":"2015"},{"key":"pone.0322490.ref004","unstructured":"Cardino PC, Chua JL, Llaga JR. Advance scorecard for boxing: Combat sport analysis with deep learning. Available at: https:\/\/animorepository.dlsu.edu.ph. Published 2022."},{"issue":"2","key":"pone.0322490.ref005","first-page":"356","article-title":"Technical and tactical aspects that differentiate winning and losing performances in boxing","volume":"11","author":"SE Ashker","year":"2011","journal-title":"Int J Perform Analysis in Sport"},{"key":"pone.0322490.ref006","first-page":"97","article-title":"Punch Types and Range Estimation in Boxing Bouts using IMU Sensors.","author":"S Manoharan","year":"2023"},{"key":"pone.0322490.ref007","unstructured":"Monfared S. Contributing factors to punching power in Boxing: A narrative review summarizing determinant factors of punching power in boxing and means of improving them. Published 2021. Available from: https:\/\/diva-portal."},{"issue":"2","key":"pone.0322490.ref008","doi-asserted-by":"crossref","first-page":"360","DOI":"10.3390\/iot1020021","article-title":"An Evaluation of Wearable Inertial Sensor Configuration and Supervised Machine Learning Models for Automatic Punch Classification in Boxing","volume":"1","author":"MTO Worsey","year":"2020","journal-title":"IoT"},{"key":"pone.0322490.ref009","doi-asserted-by":"crossref","unstructured":"Wada T, Nagahara R, Gleadhill S, Ishizuka T, Ohnuma H, Ohgi Y. Measurement of pelvic orientation angles during sprinting using a single inertial sensor. In Proceedings 2020 Jun 15; 49:1:10.","DOI":"10.3390\/proceedings2020049010"},{"key":"pone.0322490.ref010","first-page":"23","article-title":"Biomechanical characteristics of amateur boxers","volume":"15","author":"K Bu\u015bko","year":"2019","journal-title":"Archives of Budo"},{"issue":"1","key":"pone.0322490.ref011","doi-asserted-by":"crossref","DOI":"10.2174\/1874120701913010151","article-title":"A development machine for measuring the precision and the response time of punches. The","volume":"13","author":"T Jaysrichai","year":"2019","journal-title":"Open Biomedical Engineering Journal"},{"key":"pone.0322490.ref012","unstructured":"Sosopoulos K, Woldu MT. IoT smart athletics: Boxing glove sensors implementing machine learning for an integrated training solution. Published 2021.Available from: https:\/\/diva-portal."},{"issue":"3","key":"pone.0322490.ref013","doi-asserted-by":"crossref","first-page":"1223","DOI":"10.3390\/app11031223","article-title":"Application of an Artificial Neural Network to Automate the Measurement of Kinematic Characteristics of Punches in Boxing","volume":"11","author":"I Khasanshin","year":"2021","journal-title":"Applied Sciences"},{"issue":"7","key":"pone.0322490.ref014","doi-asserted-by":"crossref","first-page":"131","DOI":"10.3390\/sports11070131","article-title":"Profiling the Physical Performance of Young Boxers with Unsupervised Machine Learning: A Cross-Sectional Study","volume":"11","author":"R Merlo","year":"2023","journal-title":"Sports (Basel)"},{"issue":"23","key":"pone.0322490.ref015","doi-asserted-by":"crossref","first-page":"7882","DOI":"10.3390\/s21237882","article-title":"Application of a Validated Innovative Smart Wearable for Performance Analysis by Experienced and Non-Experienced Athletes in Boxing","volume":"21","author":"T Menzel","year":"2021","journal-title":"Sensors (Basel)"},{"key":"pone.0322490.ref016","doi-asserted-by":"crossref","first-page":"143","DOI":"10.1016\/j.cviu.2017.04.007","article-title":"Fine-grained action recognition of boxing punches from depth imagery","volume":"159","author":"S Kasiri","year":"2017","journal-title":"Computer Vision and Image Understanding"},{"issue":"4","key":"pone.0322490.ref017","first-page":"2538","article-title":"Kinematic and kinetic analysis of throwing a straight punch: the role of trunk rotation in delivering a powerful straight punch","volume":"17","author":"R Tong-Iam","year":"2017","journal-title":"Journal of Physical Education and Sport"},{"key":"pone.0322490.ref018","first-page":"25","volume-title":"Detecting clashes in boxing. InProceedings of the 3rd Polish Conference on Artificial Intelligence, Gdynia, Poland","author":"P Stefanski","year":"2022"},{"key":"pone.0322490.ref019","first-page":"165","article-title":"Augmenting the boxing game with smartphone imu-based classification system on waist.","author":"F Gu","year":"2022"},{"issue":"1","key":"pone.0322490.ref020","first-page":"47","article-title":"Measuring the force of punches and kicks among combat sport athletes using a modified punching bag with an embedded accelerometer","volume":"18","author":"K Bu\u015bko","year":"2016","journal-title":"Acta Bioeng Biomech"},{"issue":"21","key":"pone.0322490.ref021","doi-asserted-by":"crossref","first-page":"6947","DOI":"10.3390\/s21216947","article-title":"Validation of a Unique Boxing Monitoring System","volume":"21","author":"T Menzel","year":"2021","journal-title":"Sensors (Basel)"},{"key":"pone.0322490.ref022","doi-asserted-by":"crossref","first-page":"96","DOI":"10.1109\/BigDataService55688.2022.00022","article-title":"Active learning for node classification using a convex optimization approach","author":"D Agarwal","year":"2022","journal-title":"In 2022 IEEE Eighth International Conference on Big Data Computing Service and Applications (BigDataService)"},{"key":"pone.0322490.ref023","doi-asserted-by":"crossref","first-page":"557","DOI":"10.1109\/WF-IoT51360.2021.9595397","article-title":"Addressing Uncertainties within Active Learning for Industrial IoT","author":"D Agarwal","year":"2021","journal-title":"2021 IEEE 7th World Forum on Internet of Things (WF-IoT)"},{"key":"pone.0322490.ref024","article-title":"Active learning literature survey","author":"B Settles","year":"2009"},{"issue":"16","key":"pone.0322490.ref025","doi-asserted-by":"crossref","first-page":"9073","DOI":"10.3390\/app13169073","article-title":"Smart Boxing Glove \u201cRD \u03b1\u201d: IMU Combined with Force Sensor for Highly Accurate Technique and Target Recognition Using Machine Learning","volume":"13","author":"D Cizmic","year":"2023","journal-title":"Applied Sciences"},{"issue":"24","key":"pone.0322490.ref026","doi-asserted-by":"crossref","first-page":"8409","DOI":"10.3390\/s21248409","article-title":"Fusion Models for Generalized Classification of Multi-Axial Human Movement: Validation in Sport Performance","volume":"21","author":"R Amerineni","year":"2021","journal-title":"Sensors (Basel)"}],"container-title":["PLOS One"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0322490","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,5,8]],"date-time":"2025-05-08T17:37:04Z","timestamp":1746725824000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pone.0322490"}},"subtitle":[],"editor":[{"given":"Xiaohui","family":"Zhang,","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2025,5,8]]},"references-count":26,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2025,5,8]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pone.0322490","relation":{},"ISSN":["1932-6203"],"issn-type":[{"value":"1932-6203","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,5,8]]}}}