Answers are grounded in this paper and cite their sources.
OBJECTIVES: Wushu jumping kicks require highly coordinated whole-body motion during take-off and aerial execution, while landing imposes substantial biomechanical stress on the lower limbs. This study aimed to develop an integrated framework for performance and landing-safety assessment. METHODS: The OpenPose human pose-estimation framework, using a VGG-19 convolutional backbone for feature extraction, detected anatomical joints in 2D video frames. These 2D joint coordinates were subsequently reconstructed into 3D poses using individualized anthropometric segment lengths, body proportions, anatomical constraints, and temporal continuity across sequential frames. RESULTS: Reconstructed 3D models showed a proximal-to-distal energy-transfer sequence, with joint angular velocities peaking from 420.5 to 940.8°/s and time lags shortening from 122 to 56 ms. Landing analysis showed a 44.3° ± 2.5°foot angle, deep knee flexion (63.2°-108.5°) for shock attenuation, and asymmetry below the 10 % risk threshold. Stabilization was achieved within 0.85 s, while VR enabled detailed visualization of joint trajectories, movement sequencing, and landing asymmetries. CONCLUSIONS: The proposed marker-free framework integrates deep-learning-based 3D pose estimation, biomechanical analysis, and VR visualization to provide a comprehensive tool for evaluating Wushu kicking performance and landing safety.
No takes yet. Share an insight, caveat, or question.
Xie et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: