Yao Guo received the B.S. and M.S. degrees in electrical engineering from Sun Yat-sen University, Guangzhou, China, in 2011 and 2014, respectively. He obtained a Ph.D. degree in Mechanical and Biomedical Engineering from the City University of Hong Kong, Hong Kong, in 2018. He is currently a Research Associate with the Hamlyn Centre for Robotic Surgery, Imperial College London, London. His research interests include robot vision, machine learning, gait analysis, assistive robotics, human-robot interaction, and human body kinematics and biomechanics.
3D gait analysis is a powerful tool for detecting and monitoring movement dysfunctions in the elderly and patients with neurological disorders and musculoskeletal problems. Here we present a general framework for 3D gait monitoring with a single RGB-D camera, which consists of the localization and tracking of the 3D human skeleton, followed by an abnormal gait pattern classification. Firstly, both the 6D camera pose and the 3D lower limb skeleton are real-time tracked in a canonical coordinate system based on Simultaneously Localization and Mapping (SLAM) and the state-of-the-art human pose estimation. Next, to tackle the low-quality of depth data, we present a coupled real-synthetic domain adaptation method that enables domain transfer between high-quality synthetic depth data and real RGB-D camera information for super-resolution depth recovery. To further improve the recognition accuracy, a disentangle feature learning method is introduced to extract subject-invariant gait features, hence, the trained classifier can be well adapted to new subjects. To evaluate the robustness of the system, we collected multi-cameras, ground truth data from sixteen healthy volunteers performing five gait patterns that mimic common gait abnormalities. The experiment results demonstrate that our proposed system can achieve good lower limb pose estimation and superior recognition accuracy compared to previous methods.
Institute of Medical Robotics