Learning Dynamics from Kinematics: Estimating 2D Foot Pressure Maps from Video Frames
This work addresses stability analysis for applications in kinesiology, medicine, and robotics, but it is incremental as it applies deep learning to a specific domain task.
The paper tackled the problem of estimating foot pressure maps and Center of Pressure (CoP) from 2D human pose in video to aid stability analysis, achieving significantly more accurate results than a baseline K-Nearest Neighbors method and meeting lab-based measurement expectations.
Pose stability analysis is the key to understanding locomotion and control of body equilibrium, with applications in numerous fields such as kinesiology, medicine, and robotics. In biomechanics, Center of Pressure (CoP) is used in studies of human postural control and gait. We propose and validate a novel approach to learn CoP from pose of a human body to aid stability analysis. More specifically, we propose an end-to-end deep learning architecture to regress foot pressure heatmaps, and hence the CoP locations, from 2D human pose derived from video. We have collected a set of long (5min +) choreographed Taiji (Tai Chi) sequences of multiple subjects with synchronized foot pressure and video data. The derived human pose data and corresponding foot pressure maps are used jointly in training a convolutional neural network with residual architecture, named PressNET. Cross-subject validation results show promising performance of PressNET, significantly outperforming the baseline method of K-Nearest Neighbors. Furthermore, we demonstrate that our computation of center of pressure (CoP) from PressNET is not only significantly more accurate than those obtained from the baseline approach but also meets the expectations of corresponding lab-based measurements of stability studies in kinesiology.