Details
| Original language | English |
|---|---|
| Title of host publication | 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) |
| Pages | 8856-8865 |
| Number of pages | 10 |
| ISBN (electronic) | 978-1-6654-6946-3 |
| Publication status | Published - 2022 |
| Event | 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022 - New Orleans, United States Duration: 19 Jun 2022 → 24 Jun 2022 |
Publication series
| Name | Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition |
|---|---|
| Publisher | IEEE Computer Society |
| ISSN (Print) | 1063-6919 |
Abstract
Multi-Camera Multi-Object Tracking is currently drawing attention in the computer vision field due to its superior performance in real-world applications such as video surveillance with crowded scenes or in wide spaces. In this work, we propose a mathematically elegant multi-camera multiple object tracking approach based on a spatial-temporal lifted multicut formulation. Our model utilizes state-of-the-art tracklets produced by single-camera trackers as proposals. As these tracklets may contain ID-Switch errors, we refine them through a novel pre-clustering obtained from 3D geometry projections. As a result, we derive a better tracking graph without ID switches and more precise affinity costs for the data association phase. Tracklets are then matched to multi-camera trajectories by solving a global lifted multicut formulation that incorporates short and long-range temporal interactions on tracklets located in the same camera as well as inter-camera ones. Experimental results on the WildTrack dataset yield near-perfect performance, outperforming state-of-the-art trackers on Campus while being on par on the PETS-09 dataset. We will release our implementations at this link https://github.com/nhmduy/LMGP.
Keywords
- Motion and tracking, Optimization methods, Scene analysis and understanding, Video analysis and understanding, Vision applications and systems
ASJC Scopus subject areas
- Computer Science(all)
- Software
- Computer Science(all)
- Computer Vision and Pattern Recognition
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2022. p. 8856-8865 (Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition).
Research output: Chapter in book/report/conference proceeding › Conference contribution › Research › peer review
}
TY - GEN
T1 - LMGP
T2 - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
AU - Nguyen, Duy M.H.
AU - Henschel, Roberto
AU - Rosenhahn, Bodo
AU - Sonntag, Daniel
AU - Swoboda, Paul
N1 - Funding Information: This research is sponsored by the XAINES and pAItient projects (BMBF grant no. 01IW20005 and BMG grant no. 2520DAT0P2 respectively).
PY - 2022
Y1 - 2022
N2 - Multi-Camera Multi-Object Tracking is currently drawing attention in the computer vision field due to its superior performance in real-world applications such as video surveillance with crowded scenes or in wide spaces. In this work, we propose a mathematically elegant multi-camera multiple object tracking approach based on a spatial-temporal lifted multicut formulation. Our model utilizes state-of-the-art tracklets produced by single-camera trackers as proposals. As these tracklets may contain ID-Switch errors, we refine them through a novel pre-clustering obtained from 3D geometry projections. As a result, we derive a better tracking graph without ID switches and more precise affinity costs for the data association phase. Tracklets are then matched to multi-camera trajectories by solving a global lifted multicut formulation that incorporates short and long-range temporal interactions on tracklets located in the same camera as well as inter-camera ones. Experimental results on the WildTrack dataset yield near-perfect performance, outperforming state-of-the-art trackers on Campus while being on par on the PETS-09 dataset. We will release our implementations at this link https://github.com/nhmduy/LMGP.
AB - Multi-Camera Multi-Object Tracking is currently drawing attention in the computer vision field due to its superior performance in real-world applications such as video surveillance with crowded scenes or in wide spaces. In this work, we propose a mathematically elegant multi-camera multiple object tracking approach based on a spatial-temporal lifted multicut formulation. Our model utilizes state-of-the-art tracklets produced by single-camera trackers as proposals. As these tracklets may contain ID-Switch errors, we refine them through a novel pre-clustering obtained from 3D geometry projections. As a result, we derive a better tracking graph without ID switches and more precise affinity costs for the data association phase. Tracklets are then matched to multi-camera trajectories by solving a global lifted multicut formulation that incorporates short and long-range temporal interactions on tracklets located in the same camera as well as inter-camera ones. Experimental results on the WildTrack dataset yield near-perfect performance, outperforming state-of-the-art trackers on Campus while being on par on the PETS-09 dataset. We will release our implementations at this link https://github.com/nhmduy/LMGP.
KW - Motion and tracking
KW - Optimization methods
KW - Scene analysis and understanding
KW - Video analysis and understanding
KW - Vision applications and systems
UR - http://www.scopus.com/inward/record.url?scp=85138289826&partnerID=8YFLogxK
U2 - 10.48550/arXiv.2111.11892
DO - 10.48550/arXiv.2111.11892
M3 - Conference contribution
AN - SCOPUS:85138289826
SN - 978-1-6654-6947-0
T3 - Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 8856
EP - 8865
BT - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Y2 - 19 June 2022 through 24 June 2022
ER -