3D Human pose estimation using geometric self-supervision with temporal methods

dc.contributor.authorBau, Nandi
dc.contributor.supervisorKlein, Richard
dc.date.accessioned2025-06-09T16:08:43Z
dc.date.issued2024-09
dc.descriptionA dissertation submitted in fulfilment of the requirements for the degree of Master of Science, to the Faculty of Science, School of Computer Science & Applied Mathematics, University of the Witwatersrand, Johannesburg, 2024
dc.description.abstractThis dissertation explores the enhancement of 3D human pose estimation (HPE) through self-supervised learning methods that reduce reliance on heavily annotated datasets. Recognising the limitations of data acquired in controlled lab settings, the research investigates the potential of geometric self-supervision combined with temporal information to improve model performance in real-world scenarios. A Temporal Dilated Convolutional Network (TDCN) model, employing Kalman filter post-processing, is proposed and evaluated on both ground-truth and in-the-wild data from the Human3.6M dataset. The results demonstrate a competitive Mean Per Joint Position Error (MPJPE) of 62.09mm on unseen data, indicating a promising direction for self-supervised learning in 3D HPE and suggesting a viable pathway towards reducing the gap with fully supervised methods. This study underscores the value of self-supervised temporal dynamics in advancing pose estimation techniques, potentially making them more accessible and broadly applicable in real-world applications.
dc.description.sponsorshipDSI-NRF Centre of Excellence in Mathematical and Statistical Sciences (CoE-MaSS), South Africa
dc.description.submitterMMM2025
dc.facultyFaculty of Science
dc.identifier000-0002-2067-418X
dc.identifier.citationBau, Nandi. (2024). 3D Human pose estimation using geometric self-supervision with temporal methods. [Masters' dissertation, University of the Witwatersrand, Johannesburg]. WIReDSpace. https://hdl.handle.net/10539/45085
dc.identifier.urihttps://hdl.handle.net/10539/45085
dc.language.isoen
dc.publisherUniversity of the Witwatersrand, Johannesburg
dc.rights©2024 University of the Witwatersrand, Johannesburg. All rights reserved. The copyright in this work vests in the University of the Witwatersrand, Johannesburg. No part of this work may be reproduced or transmitted in any form or by any means, without the prior written permission of University of the Witwatersrand, Johannesburg.
dc.rights.holderUniversity of the Witwatersrand, Johannesburg
dc.schoolSchool of Computer Science and Applied Mathematics
dc.subjectMachine learning
dc.subjectApplied machine learning
dc.subjectComputer vision
dc.subjectHuman pose estimation
dc.subject3D human pose estimation
dc.subjectGeometric self-supervision
dc.subjectTemporal methods
dc.subjectTemporal dilated convolutional neural networks
dc.subjectDeep learning
dc.subjectUCTD
dc.subject.primarysdgSDG-9: Industry, innovation and infrastructure
dc.subject.secondarysdgSDG-4: Quality education
dc.title3D Human pose estimation using geometric self-supervision with temporal methods
dc.typeDissertation

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
Bau_3D Human_2024.pdf
Size:
2.99 MB
Format:
Adobe Portable Document Format

License bundle

Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
2.43 KB
Format:
Item-specific license agreed upon to submission
Description: