Siamese Contour Segmentation Network for Multi-State Object Tracking
Tracking multiple states of the target simultaneously is currently a research hotspot in object tracking. Existing methods obtain the initial bounding box with multi-scale search, anchor-based regression, or anchor-free regression. Then, each pixel within bounding box is classified and the mask of t...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10310210/ |
| _version_ | 1797545239327539200 |
|---|---|
| author | Hao Li Ziwen Sun Chong Ling Chao Xu |
| author_facet | Hao Li Ziwen Sun Chong Ling Chao Xu |
| author_sort | Hao Li |
| collection | DOAJ |
| description | Tracking multiple states of the target simultaneously is currently a research hotspot in object tracking. Existing methods obtain the initial bounding box with multi-scale search, anchor-based regression, or anchor-free regression. Then, each pixel within bounding box is classified and the mask of target is used to fit a fine bounding box for accurate tracking. However, their computation complexity is high and the accuracy of mask and fine bounding box is limited by initial bounding box. Based on SiamMask, the Siamese contour segmentation network (SiamCS) is proposed for multi-state object tracking to address these issues. This end-to-end network eliminates need to pre-define anchor based on prior knowledge, and reduces hyperparameters. With SiamCS, multi-state object tracking method formulates object tracking as region classification and contour regression to obtain bounding box, contour, and mask of target at the same time. Moreover, according to geometric meaning of definite integral and difficulty of sample, difficulty sensitive contour-intersection over union loss function is proposed to solve the problem of independent regression of contour parameters. Extensive experiments on OTB100 (92.5%Precision), UAV123 (83.5%Precision), LaSOT (64.8%AUC), TrackingNet (81.5%AUC), GOT-10K (66.3%AO), and VOT2020 (53.5%EAO) show that SiamCS outperforms many state-of-the-art trackers and achieves leading performance. |
| first_indexed | 2024-03-10T14:12:41Z |
| format | Article |
| id | doaj.art-65c1f62e3de14e098ac874353e0d98b8 |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-03-10T14:12:41Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-65c1f62e3de14e098ac874353e0d98b82023-11-21T00:01:00ZengIEEEIEEE Access2169-35362023-01-011112663412664210.1109/ACCESS.2023.333092410310210Siamese Contour Segmentation Network for Multi-State Object TrackingHao Li0https://orcid.org/0009-0009-2668-7908Ziwen Sun1https://orcid.org/0009-0001-1714-8939Chong Ling2https://orcid.org/0009-0001-6503-7715Chao Xu3https://orcid.org/0000-0002-3417-6464Unit 75220 of PLA, Huizhou, ChinaArmy Artillery and Air Defense Academy of PLA, Hefei, ChinaArmy Artillery and Air Defense Academy of PLA, Hefei, ChinaArmy Artillery and Air Defense Academy of PLA, Hefei, ChinaTracking multiple states of the target simultaneously is currently a research hotspot in object tracking. Existing methods obtain the initial bounding box with multi-scale search, anchor-based regression, or anchor-free regression. Then, each pixel within bounding box is classified and the mask of target is used to fit a fine bounding box for accurate tracking. However, their computation complexity is high and the accuracy of mask and fine bounding box is limited by initial bounding box. Based on SiamMask, the Siamese contour segmentation network (SiamCS) is proposed for multi-state object tracking to address these issues. This end-to-end network eliminates need to pre-define anchor based on prior knowledge, and reduces hyperparameters. With SiamCS, multi-state object tracking method formulates object tracking as region classification and contour regression to obtain bounding box, contour, and mask of target at the same time. Moreover, according to geometric meaning of definite integral and difficulty of sample, difficulty sensitive contour-intersection over union loss function is proposed to solve the problem of independent regression of contour parameters. Extensive experiments on OTB100 (92.5%Precision), UAV123 (83.5%Precision), LaSOT (64.8%AUC), TrackingNet (81.5%AUC), GOT-10K (66.3%AO), and VOT2020 (53.5%EAO) show that SiamCS outperforms many state-of-the-art trackers and achieves leading performance.https://ieeexplore.ieee.org/document/10310210/Object trackingSiamese networkvideo object segmentationcontour segmentationmultiple target states |
| spellingShingle | Hao Li Ziwen Sun Chong Ling Chao Xu Siamese Contour Segmentation Network for Multi-State Object Tracking IEEE Access Object tracking Siamese network video object segmentation contour segmentation multiple target states |
| title | Siamese Contour Segmentation Network for Multi-State Object Tracking |
| title_full | Siamese Contour Segmentation Network for Multi-State Object Tracking |
| title_fullStr | Siamese Contour Segmentation Network for Multi-State Object Tracking |
| title_full_unstemmed | Siamese Contour Segmentation Network for Multi-State Object Tracking |
| title_short | Siamese Contour Segmentation Network for Multi-State Object Tracking |
| title_sort | siamese contour segmentation network for multi state object tracking |
| topic | Object tracking Siamese network video object segmentation contour segmentation multiple target states |
| url | https://ieeexplore.ieee.org/document/10310210/ |
| work_keys_str_mv | AT haoli siamesecontoursegmentationnetworkformultistateobjecttracking AT ziwensun siamesecontoursegmentationnetworkformultistateobjecttracking AT chongling siamesecontoursegmentationnetworkformultistateobjecttracking AT chaoxu siamesecontoursegmentationnetworkformultistateobjecttracking |