Related Experiment Video
Updated: Sep 8, 2025
08:25
Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
Published on: May 7, 2019
9.1K
An overview of Human Action Recognition in sports based on Computer Vision
Kristina Host1, Marina Ivašić-Kos1
1Faculty of Informatics and Digital Technologies; Center for Artificial Intelligence and Cybersecurity, University of Rijeka, Radmile Matejčić 2, Rijeka, 51000, Croatia.
Heliyon
|June 16, 2022
Summary
Human Action Recognition (HAR) in sports uses Computer Vision to identify players and their actions. This paper systematizes actions by complexity and reviews HAR methods and datasets for sports applications.
Area of Science:
- Computer Vision
- Sports Science
- Artificial Intelligence
Background:
- Human Action Recognition (HAR) is crucial for analyzing sports activities like volleyball, basketball, soccer, and tennis.
- Current HAR research often focuses on everyday actions, but its application in sports is growing.
- Recognizing player actions during training, matches, or competitions is vital for performance monitoring and statistical analysis.
Purpose of the Study:
- To provide an overview of Human Action Recognition applications specifically within the sports domain.
- To propose a novel systematization of sports actions based on complexity, performance level, and interactions.
- To highlight the role of Computer Vision in sports HAR and review relevant datasets.
Main Methods:
- Review of existing Human Action Recognition methods applied to sports.
- Analysis of publicly available datasets for sports action recognition.
- Systematization of sports actions considering complexity and interaction levels.
Main Results:
- Computer Vision is the primary approach for HAR in sports.
- A new framework for classifying sports actions based on complexity is proposed.
- Key datasets for sports HAR research are identified and discussed.
Conclusions:
- HAR is increasingly tailored towards specific sports domains, moving beyond general human actions.
- The proposed systematization aids in understanding and analyzing the complexity of sports actions.
- This work provides a valuable resource for researchers focusing on Computer Vision-based HAR in sports.
Keywords:
Action systematizationHuman Action RecognitionHuman action recognition in sportsMachine learningSportSports dataset
