Simple temporal interpolation improves missing joint data in gait analysis
Recovering Biomechanical Signals from Missing Keypoints Using Temporal Interpolation in Monocular Gait Analysis
Computer Vision and Pattern Recognition
Summary
Gait analysis with a single camera can miss important body points like ankles because of obstacles or errors. The authors show that a very simple method, interpolating data over time, can almost perfectly fill in missing ankle positions. This method made knee angle measurements much more accurate without needing complicated computer models. Their findings suggest that quick and easy fixes can work well for tracking walking in less-than-ideal conditions.
What this means in practice
- •For mobile health app developers: Implement real-time gait monitoring on smartphones even when some foot landmarks are temporarily invisible or undetected.
- •For sports biomechanics teams: Improve accuracy of joint angle measurements during athlete motion capture when low-cost or single-camera setups miss foot markers.
Authors
Shubham Jariwala
Abstract
Monocular pose estimation enables low-cost gait analysis but is sensitive to missing keypoints caused by occlusion, detection errors, or efficiency-driven model reduction. While prior work on recovering missing joints focuses on complex learned models, the effectiveness of simple temporal methods remains underexplored. We evaluate knee-angle estimation under a missing-ankle-keypoint condition and test a first-order temporal interpolation scheme as a recovery mechanism. Across 527 frames of monocular walking video (428 with valid baseline detections), removing the ankle keypoint increased mean angular error to 23.4° +/- 46.7° and collapsed signal variance to near zero. Temporal interpolation reduced error to 1.1° +/- 6.7° and restored variance and smoothness to within a few percent of baseline. These results indicate that gait signals possess sufficient temporal redundancy for a simple, computationally trivial interpolation scheme to recover a critical missing joint, without resorting to learned reconstruction models. The findings support low-complexity, real-time-compatible designs for gait analysis in resource-constrained or occlusion-prone monocular settings.