
Localizing Anthropometric Landmarks Using 3-D Surface Features - 19.074
C. Shu et al., "Localizing Anthropometric Landmarks Using 3-D Surface Features", in Proc. of 3DBODY.TECH 2019 - 10th Int. Conf. and Exh. on 3D Body Scanning and Processing Technologies, Lugano, Switzerland, 22-23 Oct. 2019, pp. 74-80, doi:10.15221/19.074.
Title:
Localizing Anthropometric Landmarks Using 3-D Surface Features
Authors:
Chang SHU 1, Pengcheng XI 1, Allan KEEFE 2
1 National Research Council of Canada, Ottawa, Canada;
2 Defence Research and Development Canada, Toronto, Canada
Abstract:
Accurate localization of anthropometric landmarks is crucial for processing and analyzing 3-D anthropometric data. For example, landmarks are used to extract dimensional measurements from 3-D scans of human bodies. They can also be used to fit a template model to the scans such that a correspondence across the scans can be established. From this correspondence, we can perform statistical shape analysis to understand the variabilities of human shapes. In this paper, we propose a new method for localizing anthropometric landmarks using a combination of 3-D surface features and the latest deep learning techniques. The method makes use of geometric features represented as descriptor vectors. We first identify a set of locations that exhibit salient geometric features. Then we use pre-registered 3-D models to train a classifier for each geometric landmark. With the geometric landmarks, we fit a template to the data scan. The full set of anthropometric landmarks can be predicted from the template-fitted model. We validate our method using the 2012 Canadian Forces Anthropometric Survey (CFAS) dataset where 2,200 full-body scans were collected.
Details:
Full paper: 19074shu.pdf
Proceedings: 3DBODY.TECH 2019, 22-23 Oct. 2019, Lugano, Switzerland
Pages: 74-80
DOI: 10.15221/19.074
License/Copyright notice:
Proceedings: © Hometrica Consulting - Dr. Nicola D'Apuzzo, Switzerland, hometrica.ch.
Authors retain all rights to individual papers, which are licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
The papers appearing in the proceedings reflect the author's opinions. Their inclusion in the proceedings does not necessary constitute endorsement by the editor or by the publisher.
Note: click the + on the top left of the page to open/close the menu.