A unified multi-view multi-person tracking framework
Abstract Despite significant developments in 3D multi-view multi-person (3D MM) tracking, current frameworks separately target footprint tracking, or pose tracking. Frameworks designed for the former cannot be used for the latter, because they directly obtain 3D positions on the ground plane via a h...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-11-01
|
| Series: | Computational Visual Media |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s41095-023-0334-8 |
| _version_ | 1797362999740071936 |
|---|---|
| author | Fan Yang Shigeyuki Odashima Sosuke Yamao Hiroaki Fujimoto Shoichi Masui Shan Jiang |
| author_facet | Fan Yang Shigeyuki Odashima Sosuke Yamao Hiroaki Fujimoto Shoichi Masui Shan Jiang |
| author_sort | Fan Yang |
| collection | DOAJ |
| description | Abstract Despite significant developments in 3D multi-view multi-person (3D MM) tracking, current frameworks separately target footprint tracking, or pose tracking. Frameworks designed for the former cannot be used for the latter, because they directly obtain 3D positions on the ground plane via a homography projection, which is inapplicable to 3D poses above the ground. In contrast, frameworks designed for pose tracking generally isolate multi-view and multi-frame associations and may not be sufficiently robust for footprint tracking, which utilizes fewer key points than pose tracking, weakening multi-view association cues in a single frame. This study presents a unified multi-view multi-person tracking framework to bridge the gap between footprint tracking and pose tracking. Without additional modifications, the framework can adopt monocular 2D bounding boxes and 2D poses as its input to produce robust 3D trajectories for multiple persons. Importantly, multi-frame and multi-view information are jointly employed to improve association and triangulation. Our framework is shown to provide state-of-the-art performance on the Campus and Shelf datasets for 3D pose tracking, with comparable results on the WILDTRACK and MMPTRACK datasets for 3D footprint tracking. |
| first_indexed | 2024-03-08T16:15:18Z |
| format | Article |
| id | doaj.art-72ef66a837574b56b534bc3228dc1c23 |
| institution | Directory Open Access Journal |
| issn | 2096-0433 2096-0662 |
| language | English |
| last_indexed | 2024-03-08T16:15:18Z |
| publishDate | 2023-11-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Computational Visual Media |
| spelling | doaj.art-72ef66a837574b56b534bc3228dc1c232024-01-07T12:39:04ZengSpringerOpenComputational Visual Media2096-04332096-06622023-11-0110113716010.1007/s41095-023-0334-8A unified multi-view multi-person tracking frameworkFan Yang0Shigeyuki Odashima1Sosuke Yamao2Hiroaki Fujimoto3Shoichi Masui4Shan Jiang5Fujitsu ResearchFujitsu ResearchFujitsu ResearchFujitsu ResearchFujitsu ResearchFujitsu ResearchAbstract Despite significant developments in 3D multi-view multi-person (3D MM) tracking, current frameworks separately target footprint tracking, or pose tracking. Frameworks designed for the former cannot be used for the latter, because they directly obtain 3D positions on the ground plane via a homography projection, which is inapplicable to 3D poses above the ground. In contrast, frameworks designed for pose tracking generally isolate multi-view and multi-frame associations and may not be sufficiently robust for footprint tracking, which utilizes fewer key points than pose tracking, weakening multi-view association cues in a single frame. This study presents a unified multi-view multi-person tracking framework to bridge the gap between footprint tracking and pose tracking. Without additional modifications, the framework can adopt monocular 2D bounding boxes and 2D poses as its input to produce robust 3D trajectories for multiple persons. Importantly, multi-frame and multi-view information are jointly employed to improve association and triangulation. Our framework is shown to provide state-of-the-art performance on the Campus and Shelf datasets for 3D pose tracking, with comparable results on the WILDTRACK and MMPTRACK datasets for 3D footprint tracking.https://doi.org/10.1007/s41095-023-0334-8multi-camera multi-person trackingpose trackingfootprint trackingtriangulationspatiotemporal clustering |
| spellingShingle | Fan Yang Shigeyuki Odashima Sosuke Yamao Hiroaki Fujimoto Shoichi Masui Shan Jiang A unified multi-view multi-person tracking framework Computational Visual Media multi-camera multi-person tracking pose tracking footprint tracking triangulation spatiotemporal clustering |
| title | A unified multi-view multi-person tracking framework |
| title_full | A unified multi-view multi-person tracking framework |
| title_fullStr | A unified multi-view multi-person tracking framework |
| title_full_unstemmed | A unified multi-view multi-person tracking framework |
| title_short | A unified multi-view multi-person tracking framework |
| title_sort | unified multi view multi person tracking framework |
| topic | multi-camera multi-person tracking pose tracking footprint tracking triangulation spatiotemporal clustering |
| url | https://doi.org/10.1007/s41095-023-0334-8 |
| work_keys_str_mv | AT fanyang aunifiedmultiviewmultipersontrackingframework AT shigeyukiodashima aunifiedmultiviewmultipersontrackingframework AT sosukeyamao aunifiedmultiviewmultipersontrackingframework AT hiroakifujimoto aunifiedmultiviewmultipersontrackingframework AT shoichimasui aunifiedmultiviewmultipersontrackingframework AT shanjiang aunifiedmultiviewmultipersontrackingframework AT fanyang unifiedmultiviewmultipersontrackingframework AT shigeyukiodashima unifiedmultiviewmultipersontrackingframework AT sosukeyamao unifiedmultiviewmultipersontrackingframework AT hiroakifujimoto unifiedmultiviewmultipersontrackingframework AT shoichimasui unifiedmultiviewmultipersontrackingframework AT shanjiang unifiedmultiviewmultipersontrackingframework |