Learning Geometry Information of Target for Visual Object Tracking with Siamese Networks
Recently, Siamese architecture has been widely used in the field of visual tracking, and has achieved great success. Most Siamese network based trackers aggregate the target information of two branches by cross-correlation. However, since the location of the sampling points in the search feature are...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-11-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/21/23/7790 |
| _version_ | 1797507229131210752 |
|---|---|
| author | Hang Chen Weiguo Zhang Danghui Yan |
| author_facet | Hang Chen Weiguo Zhang Danghui Yan |
| author_sort | Hang Chen |
| collection | DOAJ |
| description | Recently, Siamese architecture has been widely used in the field of visual tracking, and has achieved great success. Most Siamese network based trackers aggregate the target information of two branches by cross-correlation. However, since the location of the sampling points in the search feature area is pre-fixed in cross-correlation operation, these trackers suffer from either background noise influence or missing foreground information. Moreover, the cross-correlation between the template and the search area neglects the geometry information of the target. In this paper, we propose a Siamese deformable cross-correlation network to model the geometric structure of target and improve the performance of visual tracking. We propose to learn an offset field end-to-end in cross-correlation. With the guidance of the offset field, the sampling in the search image area can adapt to the deformation of the target, and realize the modeling of the geometric structure of the target. We further propose an online classification sub-network to model the variation of target appearance and enhance the robustness of the tracker. Extensive experiments are conducted on four challenging benchmarks, including OTB2015, VOT2018, VOT2019 and UAV123. The results demonstrate that our tracker achieves state-of-the-art performance. |
| first_indexed | 2024-03-10T04:45:33Z |
| format | Article |
| id | doaj.art-8e3aaee0aac948fd9a46f2bcd688066d |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T04:45:33Z |
| publishDate | 2021-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-8e3aaee0aac948fd9a46f2bcd688066d2023-11-23T02:59:09ZengMDPI AGSensors1424-82202021-11-012123779010.3390/s21237790Learning Geometry Information of Target for Visual Object Tracking with Siamese NetworksHang Chen0Weiguo Zhang1Danghui Yan2Automation College, Northwestern Polytechnical University, Xi’an 710072, ChinaAutomation College, Northwestern Polytechnical University, Xi’an 710072, ChinaAutomation College, Northwestern Polytechnical University, Xi’an 710072, ChinaRecently, Siamese architecture has been widely used in the field of visual tracking, and has achieved great success. Most Siamese network based trackers aggregate the target information of two branches by cross-correlation. However, since the location of the sampling points in the search feature area is pre-fixed in cross-correlation operation, these trackers suffer from either background noise influence or missing foreground information. Moreover, the cross-correlation between the template and the search area neglects the geometry information of the target. In this paper, we propose a Siamese deformable cross-correlation network to model the geometric structure of target and improve the performance of visual tracking. We propose to learn an offset field end-to-end in cross-correlation. With the guidance of the offset field, the sampling in the search image area can adapt to the deformation of the target, and realize the modeling of the geometric structure of the target. We further propose an online classification sub-network to model the variation of target appearance and enhance the robustness of the tracker. Extensive experiments are conducted on four challenging benchmarks, including OTB2015, VOT2018, VOT2019 and UAV123. The results demonstrate that our tracker achieves state-of-the-art performance.https://www.mdpi.com/1424-8220/21/23/7790visual object trackingdeformable convolutiondeformable cross-correlationSiamese network |
| spellingShingle | Hang Chen Weiguo Zhang Danghui Yan Learning Geometry Information of Target for Visual Object Tracking with Siamese Networks Sensors visual object tracking deformable convolution deformable cross-correlation Siamese network |
| title | Learning Geometry Information of Target for Visual Object Tracking with Siamese Networks |
| title_full | Learning Geometry Information of Target for Visual Object Tracking with Siamese Networks |
| title_fullStr | Learning Geometry Information of Target for Visual Object Tracking with Siamese Networks |
| title_full_unstemmed | Learning Geometry Information of Target for Visual Object Tracking with Siamese Networks |
| title_short | Learning Geometry Information of Target for Visual Object Tracking with Siamese Networks |
| title_sort | learning geometry information of target for visual object tracking with siamese networks |
| topic | visual object tracking deformable convolution deformable cross-correlation Siamese network |
| url | https://www.mdpi.com/1424-8220/21/23/7790 |
| work_keys_str_mv | AT hangchen learninggeometryinformationoftargetforvisualobjecttrackingwithsiamesenetworks AT weiguozhang learninggeometryinformationoftargetforvisualobjecttrackingwithsiamesenetworks AT danghuiyan learninggeometryinformationoftargetforvisualobjecttrackingwithsiamesenetworks |