Abstract
In sport science, athlete tracking and motion analysis are essential for monitoring and optimizing training programs, with the goal of increasing success in competition and preventing injury. At present, contact-free, camera-based, multi-athlete detection and tracking have become a reality, mainly due to the advances in machine learning regarding computer vision and, specifically, advances in artificial convolutional neural networks (CNN), used for human pose estimation (HPE-CNN) in image sequences. Sport science in general, as well as coaches and athletes in particular, would greatly benefit from HPE-CNN-based tracking, but the sheer amount of HPE-CNNs available, as well as their complexity, pose a hurdle to the adoption of this new technology. It is unclear how many HPE-CNNs which are available at present are ready to use in out-of-the-box inference to squash, to what extent they allow motion analysis and if detections can easily be used to provide insight to coaches and athletes. Therefore, we conducted a systematic investigation of more than 250 HPE-CNNs. After applying our selection criteria of open-source, pre-trained, state-of-the-art and ready-to-use, five variants of three HPE-CNNs remained, and were evaluated in the context of motion analysis for the racket sport of squash. Specifically, we are interested in detecting player’s feet in videos from a single camera and investigated the detection accuracy of all HPE-CNNs. To that end, we created a ground-truth dataset from publicly available squash videos by developing our own annotation tool and manually labeling frames and events. We present heatmaps, which depict the court floor using a color scale and highlight areas according to the relative time for which a player occupied that location during matchplay. These are used to provide insight into detections. Finally, we created a decision flow chart to help sport scientists, coaches and athletes to decide which HPE-CNN is best for player detection and tracking in a given application scenario.
Highlights
Training is an integral part of sports
RQ1: We found that three different human pose estimation (HPE)-convolutional neural networks (CNN) out of five variants are ready to use for out-of-the-box inference on squash data for motion analysis; RQ2: Overall, our evaluation procedure has shown sufficient accuracy for the identified HPE-CNNs on a domain-specific squash dataset; RQ3: Our heatmap visualization technique has been shown to technically be able to present detections or labels for visual assessment
We investigated the usability, accuracy, and applicability of pre-trained, state-of-the-art HPE-CNN models in detecting players’ feet in real-world squash videos
Summary
Training is an integral part of sports. Well-planned and conscientiously executed adaptation mechanisms can lead to improvements in athletes’ performance, and an optimized training program can lead to more success in competition, while decreasing the risk of injury [1]. At present, training quality and effectiveness can be quantitatively evaluated and measured using different types of sensors. For physiological core measures, such as fitness and endurance wearable sensors, monitors for heart rate, blood pressure and oxygen level are available [2]. For training aspects generally concerning movement and game tactics, motion sensors are available to measure velocity, acceleration and motion trajectories [3]. A classic example can be found in football (soccer), where team performance and collaboration is paramount and, individual player on-field locations, moves and motion paths are analyzed [4,5]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
