Article: Dynamic scene motion target segmentation method for physical education Journal: International Journal of Continuing Engineering Education and Life-Long Learning (IJCEELL) 2026 Vol.36 No.11 pp.74 - 100 Abstract: This paper proposes a segmentation algorithm integrating a spatial attention mechanism with a mask R-CNN to address safety risks in dynamic physical education scenes and improve motion target segmentation. The method first performs target segmentation using mask R-CNN, enhances feature representation through spatial attention, and completes monitoring and localisation via a visual processor and a conditional convolution instance segmentation model. Experiments show strong performance: a segmentation boundary value of 0.967, F1-score of 96.75%, and 3.27% average absolute error on the YouTube-VOS dataset, with false negative and false positive rates of 1.47% and 1.81%. On the CDnet 2014 dataset, pixel accuracy reaches 95.42% with an iteration time of 4.95 s. On a self-constructed dataset, the method achieves 0.972 average precision and 0.984 panoptic quality. These results demonstrate accurate and efficient motion target segmentation, supporting intelligent physical education applications. Inderscience Publishers - linking academia, business and industry through research

Title: Dynamic scene motion target segmentation method for physical education

Authors: Tao Lei; Yi Huang; Zhijuan Zhou

Addresses: School of Physical Education and Sports Science, Hengyang Normal University, Hengyang, 421000, China ' School of Physical Education and Sports Science, Hengyang Normal University, Hengyang, 421000, China ' School of Sports and Art, Hunan University of Medicine, Huaihua, 418000, China

Abstract: This paper proposes a segmentation algorithm integrating a spatial attention mechanism with a mask R-CNN to address safety risks in dynamic physical education scenes and improve motion target segmentation. The method first performs target segmentation using mask R-CNN, enhances feature representation through spatial attention, and completes monitoring and localisation via a visual processor and a conditional convolution instance segmentation model. Experiments show strong performance: a segmentation boundary value of 0.967, F1-score of 96.75%, and 3.27% average absolute error on the YouTube-VOS dataset, with false negative and false positive rates of 1.47% and 1.81%. On the CDnet 2014 dataset, pixel accuracy reaches 95.42% with an iteration time of 4.95 s. On a self-constructed dataset, the method achieves 0.972 average precision and 0.984 panoptic quality. These results demonstrate accurate and efficient motion target segmentation, supporting intelligent physical education applications.

Keywords: dynamic scene; motion target; physical education; spatial attention mechanism; mask region-based convolutional neural network; MRCNN; volume integral; conditional convolutions for instance segmentation; CondInst.

DOI: 10.1504/IJCEELL.2026.154664

International Journal of Continuing Engineering Education and Life-Long Learning, 2026 Vol.36 No.11, pp.74 - 100

Received: 27 Nov 2025
Accepted: 04 Feb 2026
Published online: 09 Jul 2026 *

Title: Dynamic scene motion target segmentation method for physical education

Keep up-to-date