3D Human pose estimation using geometric self-supervision with temporal methods
No Thumbnail Available
Date
2024-09
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of the Witwatersrand, Johannesburg
Abstract
This 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.
Description
A 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
Keywords
Machine learning, Applied machine learning, Computer vision, Human pose estimation, 3D human pose estimation, Geometric self-supervision, Temporal methods, Temporal dilated convolutional neural networks, Deep learning, UCTD
Citation
Bau, 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