Altmetric

Cross-domain self-supervised complete geometric representation learning for real-scanned point cloud based pathological gait analysis

File Description SizeFormat 
JBHI_20.pdfFile embargoed until 01 January 1000014.65 MBAdobe PDF    Request a copy
Title: Cross-domain self-supervised complete geometric representation learning for real-scanned point cloud based pathological gait analysis
Authors: Gu, X
Guo, Y
Yang, G-Z
Lo, B
Item Type: Journal Article
Abstract: Accurate lower-limb pose estimation is a prereq-uisite of skeleton based pathological gait analysis. To achievethis goal in free-living environments for long-term monitoring,single depth sensor has been proposed in research. However,the depth map acquired from a single viewpoint encodes onlypartial geometric information of the lower limbs and exhibitslarge variations across different viewpoints. Existing off-the-shelfthree-dimensional (3D) pose tracking algorithms and publicdatasets for depth based human pose estimation are mainlytargeted at activity recognition applications. They are relativelyinsensitive to skeleton estimation accuracy, especially at thefoot segments. Furthermore, acquiring ground truth skeletondata for detailed biomechanics analysis also requires consid-erable efforts. To address these issues, we propose a novelcross-domain self-supervised complete geometric representationlearning framework, with knowledge transfer from the unlabelledsynthetic point clouds of full lower-limb surfaces. The proposedmethod can significantly reduce the number of ground truthskeletons (with only 1%) in the training phase, meanwhileensuring accurate and precise pose estimation and capturingdiscriminative features across different pathological gait patternscompared to other methods.
Date of Acceptance: 22-Aug-2021
URI: http://hdl.handle.net/10044/1/91304
ISSN: 2168-2194
Publisher: Institute of Electrical and Electronics Engineers
Journal / Book Title: IEEE Journal of Biomedical and Health Informatics
Copyright Statement: This paper is embargoed until publication.
Publication Status: Accepted
Embargo Date: publication subject to indefinite embargo
Appears in Collections:Computing