Abstract:Automatic tagging of video recordings of sports matches and training sessions can be helpful to coaches and players and provide access to structured data at a scale that would be unfeasible if one were to rely on manual tagging. Recognition of different actions forms an essential part of sports video tagging. In this paper, the authors employ machine learning techniques to automatically recognize specific types of volleyball actions (i.e., underhand serve, overhead pass, serve, forearm pass, one hand pass, smash, and block which are manually annotated) during matches and training sessions (uncontrolled, in the wild data) based on motion data captured by inertial measurement unit sensors strapped on the wrists of eight female volleyball players. Analysis of the results suggests that all sensors in the inertial measurement unit (i.e., magnetometer, accelerometer, barometer, and gyroscope) contribute unique information in the classification of volleyball actions types. The authors demonstrate that while the accelerometer feature set provides better results than other sensors, overall (i.e., gyroscope, magnetometer, and barometer) feature fusion of the accelerometer, magnetometer, and gyroscope provides the bests results (unweighted average recall = 67.87%, unweighted average precision = 68.68%, and κ = .727), well above the chance level of 14.28%. Interestingly, it is also demonstrated that the dominant hand (unweighted average recall = 61.45%, unweighted average precision = 65.41%, and κ = .652) provides better results than the nondominant (unweighted average recall = 45.56%, unweighted average precision = 55.45, and κ = .553) hand. Apart from machine learning models, this paper also discusses a modular architecture for a system to automatically supplement video recording by detecting events of interests in volleyball matches and training sessions and to provide tailored and interactive multimodal feedback by utilizing an HTML5/JavaScript application. A proof of concept prototype developed based on this architecture is also described.

A Sport Recognition Method with Utilizing Less Motion Sensors

A Motion-based User Interface for the Control of Virtual Humans Performing Sports.

MV-Sports: A Motion and Vision Sensor Integration-Based Sports Analysis System

Quantitative Evaluation of Gymnastics Based on Multiple MEMS Sensors

The Selecting Optimal Ball-Receiving Body Parts Using Pose Sequence Analysis and Sports Biomechanics

Recognition of Badminton Shot Action Based on the Improved Hidden Markov Model

A novel approach for automatic detection and identification of inappropriate postures and movements of table tennis players

Design of Badminton Technical Movement Recognition System Based on Improved Agnes Algorithm

Action recognition for sports combined training based on wearable sensor technology and SVM prediction

A Badminton Recognition and Tracking System Based on Context Multi-feature Fusion

Design of Motor Skill Recognition and Hierarchical Evaluation System for Table Tennis Players

Towards Automatic Modeling of Volleyball Players’ Behavior for Analysis, Feedback, and Hybrid Training

Tennis Posture Classification and Recognition Based on an Improved KNN

Phase-based Quantification of Sports Performance Metrics Using a Smart IoT Sensor

A New Kinect-Based Posture Recognition Method in Physical Sports Training Based on Urban Data

Design of optical tracking sensor based on image feature extraction for badminton athlete motion recognition

Prototype Machine Learning Algorithms from Wearable Technology to Detect Tennis Stroke and Movement Actions

Transfer learning model for the motion detection of sports players

Basketball Training Posture Monitoring Based on Intelligent Wearable Device

Soft electronic material based sensor with optical network in sports application for player movement analysis using machine learning model

An IoT-Based Motion Tracking System for Next-Generation Foot-Related Sports Training and Talent Selection