An Algorithm for Standing Long Jump Distance Measurement Based on Improved YOLOv11 and Lightweight Pose Estimation

This paper proposes a vision-based algorithm for automated standing long jump (SLJ) measurement, addressing the inefficiency and inaccuracy of traditional manual methods. By integrating an enhanced YOLOv11 detector with lightweight pose estimation, the system achieves real-time performance on edge devices. Key innovations include: A FocalModulation-enhanced YOLOv11 model for precise athlete detection, improving accuracy by $1.68 \%$; A biomechanical phase recognition mechanism combining spatial-temporal analysis of heel trajectories with adaptive thresholds to identify takeoff, airborne, and landing phases; A dual-state logic control system to filter measurement noise and ensure reliable jump counting. Experimental results on 220 real-world videos demonstrate $99.44 \%$ accuracy in standard jumps, comparable to infrared systems, while outperforming conventional algorithms in complex scenarios. The system requires only smartphone cameras, eliminating dependency on specialized sensors. This work provides a cost-effective, portable solution for intelligent physical education assessment, advancing the standardization of sports training and public health monitoring.