All Keypoints You Need: Detecting Arbitrary Keypoints on the Body of Triple, High, and Long Jump Athletes
This addresses the need for detailed performance analysis for coaches in specific athletic disciplines, but it is incremental as it builds on existing human pose estimation methods.
The paper tackles the problem of fine-grained body posture analysis in triple, high, and long jump sports by proposing a method to detect arbitrary keypoints on athletes' bodies, leveraging limited annotated keypoints and auto-generated segmentation masks, with evaluations showing capability to detect keypoints on various body parts including bent joints.
Performance analyses based on videos are commonly used by coaches of athletes in various sports disciplines. In individual sports, these analyses mainly comprise the body posture. This paper focuses on the disciplines of triple, high, and long jump, which require fine-grained locations of the athlete's body. Typical human pose estimation datasets provide only a very limited set of keypoints, which is not sufficient in this case. Therefore, we propose a method to detect arbitrary keypoints on the whole body of the athlete by leveraging the limited set of annotated keypoints and auto-generated segmentation masks of body parts. Evaluations show that our model is capable of detecting keypoints on the head, torso, hands, feet, arms, and legs, including also bent elbows and knees. We analyze and compare different techniques to encode desired keypoints as the model's input and their embedding for the Transformer backbone.