Anthropometric Landmark Detection Network via Geodesic Heatmap on 3D Human Scan
In recent years, with the commercialization of three-dimensional (3D) scanners, there is an increasing demand for automated techniques that can extract anthropometric data accurately and swiftly from 3D human body scans. With advancement in computer vision and machine learning, researchers have incr...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10806647/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832592970913677312 |
|---|---|
| author | Min Hee Cha Jae Hyeon Park Ji Sun Byun Sangyeon Ahn Gyoomin Lee Seung Hyun Yoon Sung In Cho |
| author_facet | Min Hee Cha Jae Hyeon Park Ji Sun Byun Sangyeon Ahn Gyoomin Lee Seung Hyun Yoon Sung In Cho |
| author_sort | Min Hee Cha |
| collection | DOAJ |
| description | In recent years, with the commercialization of three-dimensional (3D) scanners, there is an increasing demand for automated techniques that can extract anthropometric data accurately and swiftly from 3D human body scans. With advancement in computer vision and machine learning, researchers have increasingly focused on developing automated anthropometric data extraction technique. In this paper, we propose a deep learning method for automatic anthropometric landmark extraction from 3D human scans. We adopt a coarse-to-fine approach consists of a global detection stage and a local refinement stage to fully utilize the original geometric information of input scan. Moreover, we introduce a novel geodesic heatmap that effectively captures the point distribution of 3D shapes, even in the presence of variations in scanning pose. As a result, our method provides the lowest average detection error on the SHREC’14 dataset over the six anthropometric landmarks, demonstrating a maximum error reduction of 76.14%. Additionally, we created a dataset consisting of human scans with various poses to demonstrate robustness of our method. Thanks to our new datasets, our end-to-end strategy showed its effectiveness to various human postures without any predefined features and templates. |
| format | Article |
| id | doaj-art-f5f4a184c35b44898d966b6b0f007047 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-f5f4a184c35b44898d966b6b0f0070472025-01-21T00:00:52ZengIEEEIEEE Access2169-35362024-01-011219703519704710.1109/ACCESS.2024.351967110806647Anthropometric Landmark Detection Network via Geodesic Heatmap on 3D Human ScanMin Hee Cha0https://orcid.org/0000-0003-3496-2227Jae Hyeon Park1https://orcid.org/0000-0002-6233-4394Ji Sun Byun2https://orcid.org/0009-0000-9181-9270Sangyeon Ahn3Gyoomin Lee4https://orcid.org/0009-0009-8125-9420Seung Hyun Yoon5https://orcid.org/0000-0002-0015-8305Sung In Cho6https://orcid.org/0000-0003-4251-7131Division of AI Software Convergence, Dongguk University, Seoul, Republic of KoreaDivision of AI Software Convergence, Dongguk University, Seoul, Republic of KoreaDivision of AI Software Convergence, Dongguk University, Seoul, Republic of KoreaDivision of AI Software Convergence, Dongguk University, Seoul, Republic of KoreaDivision of AI Software Convergence, Dongguk University, Seoul, Republic of KoreaDivision of AI Software Convergence, Dongguk University, Seoul, Republic of KoreaDivision of AI Software Convergence, Dongguk University, Seoul, Republic of KoreaIn recent years, with the commercialization of three-dimensional (3D) scanners, there is an increasing demand for automated techniques that can extract anthropometric data accurately and swiftly from 3D human body scans. With advancement in computer vision and machine learning, researchers have increasingly focused on developing automated anthropometric data extraction technique. In this paper, we propose a deep learning method for automatic anthropometric landmark extraction from 3D human scans. We adopt a coarse-to-fine approach consists of a global detection stage and a local refinement stage to fully utilize the original geometric information of input scan. Moreover, we introduce a novel geodesic heatmap that effectively captures the point distribution of 3D shapes, even in the presence of variations in scanning pose. As a result, our method provides the lowest average detection error on the SHREC’14 dataset over the six anthropometric landmarks, demonstrating a maximum error reduction of 76.14%. Additionally, we created a dataset consisting of human scans with various poses to demonstrate robustness of our method. Thanks to our new datasets, our end-to-end strategy showed its effectiveness to various human postures without any predefined features and templates.https://ieeexplore.ieee.org/document/10806647/Anthropometrydeep learninglandmark detection3D point cloud |
| spellingShingle | Min Hee Cha Jae Hyeon Park Ji Sun Byun Sangyeon Ahn Gyoomin Lee Seung Hyun Yoon Sung In Cho Anthropometric Landmark Detection Network via Geodesic Heatmap on 3D Human Scan IEEE Access Anthropometry deep learning landmark detection 3D point cloud |
| title | Anthropometric Landmark Detection Network via Geodesic Heatmap on 3D Human Scan |
| title_full | Anthropometric Landmark Detection Network via Geodesic Heatmap on 3D Human Scan |
| title_fullStr | Anthropometric Landmark Detection Network via Geodesic Heatmap on 3D Human Scan |
| title_full_unstemmed | Anthropometric Landmark Detection Network via Geodesic Heatmap on 3D Human Scan |
| title_short | Anthropometric Landmark Detection Network via Geodesic Heatmap on 3D Human Scan |
| title_sort | anthropometric landmark detection network via geodesic heatmap on 3d human scan |
| topic | Anthropometry deep learning landmark detection 3D point cloud |
| url | https://ieeexplore.ieee.org/document/10806647/ |
| work_keys_str_mv | AT minheecha anthropometriclandmarkdetectionnetworkviageodesicheatmapon3dhumanscan AT jaehyeonpark anthropometriclandmarkdetectionnetworkviageodesicheatmapon3dhumanscan AT jisunbyun anthropometriclandmarkdetectionnetworkviageodesicheatmapon3dhumanscan AT sangyeonahn anthropometriclandmarkdetectionnetworkviageodesicheatmapon3dhumanscan AT gyoominlee anthropometriclandmarkdetectionnetworkviageodesicheatmapon3dhumanscan AT seunghyunyoon anthropometriclandmarkdetectionnetworkviageodesicheatmapon3dhumanscan AT sungincho anthropometriclandmarkdetectionnetworkviageodesicheatmapon3dhumanscan |