Research on the Application of Computer Vision Technology in Sports Mechanics Analysis

Research on the Application of Computer Vision Technology in Sports Mechanics Analysis

Runhe Xue (Zhengzhou Health Vocational College, China)
Copyright: © 2025 |Pages: 27
DOI: 10.4018/IJeC.368010
Article PDF Download
Open access articles are freely available for download

Abstract

With the continuous development of computer vision technology and related hardware devices, motion mechanics analysis is gradually being applied in sports training. However, due to the complex characteristics of motion mechanics and irregular movements, motion mechanics analysis still has limitations in practical application scenarios. This article constructs a multi convolutional 3D CNN model that combines BN algorithm, dropout technique, and spatial pyramid pooling technique. Different features are used as inputs for 3D CNN models tested on video datasets. The experimental results show that combining the “BM+OFM+three FDF” features as model inputs can achieve high recognition accuracy. Therefore, the 3D CNN model constructed in this article can effectively improve the accuracy of motion mechanics recognition and has good application value.
Article Preview
Top

Introduction

Computer vision, an interdisciplinary research field with a long-standing history, continues to grow and evolve. Initially, its applications were predominantly centered on fundamental tasks such as object recognition, tracking, and measurement (Suo et al., 2024). However, with the integration of multiple disciplines, including image processing and pattern recognition, the scope of computer vision research has expanded significantly, accompanied by numerous challenges. Early advancements in the field were constrained by inadequate hardware and limited computational power, particularly when processing large-scale data and complex image information(Ghosh et al., 2023). These limitations in computing resources posed significant barriers to progress.

The advent of enhanced computational capabilities and the rapid growth of big data technology have created new opportunities for the field. Advances in image processing techniques now enable the efficient conversion of image data into formats conducive to deeper analysis and understanding by both humans and machines(Zhang, 2024). This development mimics the visual perception processes of the human eye and brain, effectively substituting human involvement in visual information processing. Consequently, the application boundaries of computer vision have expanded considerably. Furthermore, the introduction of deep learning technology has facilitated substantial breakthroughs in computer vision research(Wang et al., 2024). Leveraging robust feature extraction capabilities, deep learning enables automated learning and extraction of critical information from extensive datasets. This has significantly accelerated the progress of computer vision, particularly in advanced applications such as object detection, image segmentation, and action recognition.

Recently, the rapid advancement of sports and exercise science has brought increasing attention to the application of computer vision in sports mechanics analysis. This field has demonstrated significant potential in video analysis by enabling the automatic extraction and evaluation of human motion and its associated features. A central research focus in computer vision and artificial intelligence (AI) is developing methods for computers to autonomously analyze and comprehend human motion and mechanical characteristics from video data. Sports mechanics analysis not only enhances the understanding of athletic performance but also provides a scientific foundation for optimizing athlete training, monitoring health, and preventing sports-related injuries. However, the diversity of motion postures, the complexity of mechanical properties during movement, and the variability of video backgrounds present substantial challenges in motion mechanics recognition. Traditional manual feature extraction methods often depend on extensive domain expertise and involve labor-intensive processes, which undermine robustness and efficiency. Consequently, designing automated, robust, and efficient algorithms for motion mechanics recognition has emerged as a critical issue within the realm of computer vision. Such advancements are pivotal for overcoming the limitations of manual methods and advancing the precision and scalability of motion analysis systems.

Complete Article List

Search this Journal:
Reset
Volume 21: 1 Issue (2025)
Volume 20: 1 Issue (2024)
Volume 19: 7 Issues (2023)
Volume 18: 6 Issues (2022): 3 Released, 3 Forthcoming
Volume 17: 4 Issues (2021)
Volume 16: 4 Issues (2020)
Volume 15: 4 Issues (2019)
Volume 14: 4 Issues (2018)
Volume 13: 4 Issues (2017)
Volume 12: 4 Issues (2016)
Volume 11: 4 Issues (2015)
Volume 10: 4 Issues (2014)
Volume 9: 4 Issues (2013)
Volume 8: 4 Issues (2012)
Volume 7: 4 Issues (2011)
Volume 6: 4 Issues (2010)
Volume 5: 4 Issues (2009)
Volume 4: 4 Issues (2008)
Volume 3: 4 Issues (2007)
Volume 2: 4 Issues (2006)
Volume 1: 4 Issues (2005)
View Complete Journal Contents Listing