FollowMeUp Sports: New Benchmark for 2D Human Keypoint Recognition

Abstract

Human pose estimation has made significant advancement in recent years. However, the existing datasets are limited in their coverage of pose variety. In this paper, we introduce a novel benchmark "Fol-lowMeUp Sports" that makes an important advance in terms of specific postures, self-occlusion and class balance, a contribution that we feel is required for future development in human body models. This comprehensive dataset was collected using an established taxonomy of over 200 standard workout activities with three different shot angles. The collected videos cover a wider variety of specific workout activities than previous datasets including push-up, squat and body moving near the ground with severe self-occlusion or occluded by some sport equipment and outfits. Given these rich images, we perform a detailed analysis of the leading human pose estimation approaches gaining insights for the success and failures of these methods.

A Primer on Motion Capture with Deep Learning: Principles, Pitfalls, and Perspectives

“…Despite being trained on large-scale datasets of thousands of individuals, even the best architectures fail to generalize to ''atypical'' postures (with respect to the training set). This is wonderfully illustrated by the errors committed by OpenPose on yoga poses (Huang et al, 2019).…”

A Primer on Motion Capture with Deep Learning: Principles, Pitfalls and Perspectives

“…Despite being trained on large scale datasets of thousands of individuals, even the best architectures fail to generalize to "atypical" postures (with respect to the training set). This is wonderfully illustrated by the errors committed by OpenPose on yoga poses (129).…”

“…Plus, making robust networks is still a major challenge, even when trained with large amounts of data (86,172). In order to make this a possibility it will be important to develop and share common keypoint estimation benchmarks for animals as well as expand the human ones to applications of interest, such as sports (129).…”

Exploring Rare Pose in Human Pose Estimation

“…walking and playing sports) that are natural in motion. Huang et al [3] had analyzed COCO [4] dataset and reported that 85% of the dataset is composed of standing poses with the rest being either sitting or lying poses.…”