YOLOv8-Based Custom Object Detection System for Field Hockey Analysis
Article Sidebar
Main Article Content
Abstract
The objective of this study is to create custom-trained models utilizing the YOLOv8 architecture to distinguish players in fast-paced field hockey environments. The research aims to enhance performance analysis and strategic decision-making in field hockey by introducing new insights and methodologies to optimize gameplay dynamics. The methodology involved training YOLOv8 models of varying sizes (Nano, Small, Medium, Large, Extra Large) on a self-annotated dataset using a GPU, with training conducted over multiple epochs at an image size of 640. Performance metrics such as precision, recall, F-1 score, and overall accuracy were evaluated for each model variant at 100 epochs. Results indicated that YOLOv8x models achieved high precision (0.832), recall (0.861), and an overall accuracy (mAP@0.5) of 85.70% after 100 epochs, with performance varying based on model size. Additionally, confusion matrices provided detailed insights into the classification performance of YOLOv8 models, highlighting areas for improvement and strengths in object detection. This study's innovation lies in the unique application of YOLOv8-based models for object detection in field hockey, contributing to a deeper understanding of their capabilities and limitations in sports analytics, particularly in player tracking, performance analysis, and strategic decision-making during field hockey matches.
Downloads
Article Details
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
How to Cite
References
A. Li and W. Huang, “A comprehensive survey of artificial intelligence and cloud computing applications in the sports industry,” Wireless Networks, pp. 1–12, Nov. 2023, doi: 10.1007/s11276-023-03567-3.
R. Beal, T. J. Norman, and S. D. Ramchurn, “Artificial intelligence for team sports: a survey,” The Knowledge Engineering Review, vol. 34, p. e28, Dec. 2019, doi: 10.1017/S0269888919000225.
C. Direkoglu, M. Sah, and N. E. O’Connor, “Player detection in field sports,” Machine Vision and Applications, vol. 29, no. 2, pp. 187–206, Feb. 2018, doi: 10.1007/s00138-017-0893-8.
A. Chan, M. D. Levine, and M. Javan, “Player Identification in Hockey Broadcast Videos,” Expert Systems with Applications, vol. 165, p. 113891, Mar. 2021, doi: 10.1016/j.eswa.2020.113891.
S. S. A. Zaidi, M. S. Ansari, A. Aslam, N. Kanwal, M. Asghar, and B. Lee, “A survey of modern deep learning based object detection models,” Digital Signal Processing, vol. 126, p. 103514, Jun. 2022, doi: 10.1016/j.dsp.2022.103514.
J. Redmon and A. Farhadi, “YOLO9000: Better, Faster, Stronger,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, Jul. 2017, pp. 6517–6525. doi: 10.1109/CVPR.2017.690.
T. Diwan, G. Anirudh, and J. V Tembhurne, “Object detection using YOLO: challenges, architectural successors, datasets and applications,” Multimedia Tools and Applications, vol. 82, no. 6, pp. 9243–9275, Mar. 2023, doi: 10.1007/s11042-022-13644-y.
J. Liu, X. Tong, W. Li, T. Wang, Y. Zhang, and H. Wang, “Automatic player detection, labeling and tracking in broadcast soccer video,” Pattern Recognition Letters, vol. 30, no. 2, pp. 103–113, Jan. 2009, doi: 10.1016/j.patrec.2008.02.011.
V. Renò et al., “A technology platform for automatic high-level tennis game analysis,” Computer Vision and Image Understanding, vol. 159, pp. 164–175, Jun. 2017, doi: 10.1016/j.cviu.2017.01.002.
J. Han, D. Zhang, G. Cheng, N. Liu, and D. Xu, “Advanced Deep-Learning Techniques for Salient and Category-Specific Object Detection: A Survey,” IEEE Signal Processing Magazine, vol. 35, no. 1, pp. 84–100, Jan. 2018, doi: 10.1109/MSP.2017.2749125.
M. Pobar and M. Ivasic-Kos, “Active Player Detection in Handball Scenes Based on Activity Measures,” Sensors, vol. 20, no. 5, p. 1475, Mar. 2020, doi: 10.3390/s20051475.
S. H. Patel and D. Kamdar, “Object Detection in Hockey Sport Video Via Pretrained Yolov3 Based Deep Learning Model,” Ictact Journal on Image and Video Processing, vol. 13, no. FEBRUARY, pp. 2893–2898, 2023, doi: 10.21917/ijivp.2023.0412.
B. T. Naik, “Ball and Player Detection & Tracking in Soccer Videos Using Improved YOLOV3 Model,” 2021, doi: https://www.researchsquare.com/article/rs-438886/v1.
tzutalin, “tzutalin/labelImg,” GitHub. 2019. [Online]. Available: https://github.com/tzutalin/labelImg
B. Dwyer, J. Nelson, J. Solawetz, and others, “Roboflow (version 1.0)[software],” URL: https://roboflow. com. computer vision, 2022.
J. Redmon and A. Farhadi, “YOLOv3: An Incremental Improvement,” Apr. 2018, [Online]. Available: http://arxiv.org/abs/1804.02767
G. Jocher et al., “ultralytics/yolov5: v7.0 - YOLOv5 SOTA Realtime Instance Segmentation,” Nov. 2022, doi: 10.5281/ZENODO.7347926.
J. Nelson, “Train Test Split Guide and Overview.” 2020. Accessed: May 13, 2024. [Online]. Available: https://blog.roboflow.com/train-test-split-with-roboflow/
A. Mishra, “Metrics to Evaluate Your Machine Learning Algorithm.” Towards Data Science, 2018. [Online]. Available: https://towardsdatascience.com/metrics-to-evaluate-your-machine-learning-algorithm-f10ba6e38234
T. Wood, “F-Score.” 2019. [Online]. Available: https://deepai.org/machine-learning-glossary-and-terms/f-score
baeldung, “Training and Validation Loss in Deep Learning | Baeldung on Computer Science.” 2022. [Online]. Available: https://www.baeldung.com/cs/training-validation-loss-deep-learning
J. Nelson, S. Narkhede, and A. Mishra, “Understanding Confusion Matrix.” Towards Data Science, 2018. Accessed: May 13, 2024. [Online]. Available: https://towardsdatascience.com/metrics-to-evaluate-your-machine-learning-algorithm-f10ba6e38234