ABSTRACT
Sports physiotherapists and coaches are tasked with evaluating the movement quality of athletes across the spectrum of ability and experience. However, the accuracy of visual observation is low and existing technology outside of expensive lab-based solutions has limited adoption, leading to an unmet need for an efficient and accurate means to measure static and dynamic joint angles during movement, converted to movement metrics useable by practitioners. This paper proposes a set of pose landmarks for computing frequently used joint angles as metrics of interest to sports physiotherapists and coaches in assessing common strength-building human exercise movements. It then proposes a set of rules for computing these metrics for a range of common exercises (single and double drop jumps and counter-movement jumps, deadlifts and various squats) from anatomical key-points detected using video, and evaluates the accuracy of these using a published 3D human pose model trained with ground truth data derived from VICON motion capture of common rehabilitation exercises. Results show a set of mathematically defined metrics which are derived from the chosen pose landmarks, and which are sufficient to compute the metrics for each of the exercises under consideration. Comparison to ground truth data showed that root mean square angle errors were within 10° for all exercises for the following metrics: shin angle, knee varus/valgus and left/right flexion, hip flexion and pelvic tilt, trunk angle, spinal flexion lower/upper/mid and rib flare. Larger errors (though still all within 15°) were observed for shoulder flexion and ASIS asymmetry in some exercises, notably front squats and drop-jumps. In conclusion, the contribution of this paper is that a set of sufficient key-points and associated metrics for exercise assessment from 3D human pose have been uniquely defined. Further, we found generally very good accuracy of the Strided Transformer 3D pose model in predicting these metrics for the chosen set of exercises from a single mobile device camera, when trained on a suitable set of functional exercises recorded using a VICON motion capture system. Future assessment of generalization is needed.
ABSTRACT
INTRODUCTION: Single Bicycle Brashes (SBCs) are common, and underreported in official statistics. In urban environments, light rail tram tracks are a frequent factor, however, they have not yet been the subject of engineering analysis. METHOD: This study employs video-based analysis at nine Dublin city centre locations and introduces a predictive model for crossing success on tram tracks, utilising cyclist crossing angles within a Surrogate Measure of Safety (SMoS) framework. Additionally, Convolutional Neural Networks (CNNs) were explored for automatic estimation of crossing angles. RESULTS: Modelling results indicate that cyclist crossing angle is a strong predictor of crossing success, and that cyclist velocity is not. Findings also highlight the prevalence of external factors which limit crossing angles for cyclists. In particular, kerbs are a common factor, along with passing/approaching vehicles or other cyclists. Furthermore, results indicate that further training on a relatively small sample of 100 domain-specific examples can achieve substantial accuracy improvements for cyclist detection (from 0.31AP0.5 to 0.98AP0.5) and crossing angle inference from traffic camera footage. CONCLUSIONS: Ensuring safe crossing angles is important for cyclist safety around tram tracks. Infrastructural planners should aim for intuitive, self-explainable road layouts that allow for and encourage crossing angles of 60° or more - ideally 90°. PRACTICAL APPLICATIONS: The SMoS framework and the open-source SafeCross1 application offer actionable insights and tools for enhancing cyclist safety around tram tracks.