Learning Adaptive Spatial Regularization and Temporal-Aware Correlation Filters for Visual Object Tracking
Recently, discriminative correlation filters (DCF) based trackers have gained much attention and obtained remarkable achievements for their high efficiency and outstanding performance. However, undesirable boundary effects occur when the DCF-based trackers suffer from challenging situations, such as...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-11-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/10/22/4320 |
| _version_ | 1797464630730162176 |
|---|---|
| author | Liqiang Liu Tiantian Feng Yanfang Fu Chao Shen Zhijuan Hu Maoyuan Qin Xiaojun Bai Shifeng Zhao |
| author_facet | Liqiang Liu Tiantian Feng Yanfang Fu Chao Shen Zhijuan Hu Maoyuan Qin Xiaojun Bai Shifeng Zhao |
| author_sort | Liqiang Liu |
| collection | DOAJ |
| description | Recently, discriminative correlation filters (DCF) based trackers have gained much attention and obtained remarkable achievements for their high efficiency and outstanding performance. However, undesirable boundary effects occur when the DCF-based trackers suffer from challenging situations, such as occlusion, background clutters, fast motion, and so on. To address these problems, this work proposes a novel adaptive spatial regularization and temporal-aware correlation filters (ASTCF) model to deal with the boundary effects which occur in the correlation filters tracking. Firstly, our ASTCF model learns a more robust correlation filter template by introducing spatial regularization and temporal-aware components into the objective function. The adaptive spatial regularization provides a more robust appearance model to handle the large appearance changes at different times; meanwhile, the temporal-aware constraint can enhance the time continuity and consistency of this model. They make correlation filters model more discriminating, and also reduce the influence of the boundary effects during the tracking process. Secondly, the objective function can be transformed into three sub-problems with closed-form solutions and effectively solved via the alternating direction method of multipliers (ADMM). Finally, we compare our tracker with some representative methods and evaluate using three different benchmarks, including OTB2015, VOT2018 and LaSOT datasets, where the experimental results demonstrate the superiority of our tracker on most of the performance criteria compared with the existing trackers. |
| first_indexed | 2024-03-09T18:10:58Z |
| format | Article |
| id | doaj.art-89c8c561296540e8a9cd244657d486c3 |
| institution | Directory Open Access Journal |
| issn | 2227-7390 |
| language | English |
| last_indexed | 2024-03-09T18:10:58Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj.art-89c8c561296540e8a9cd244657d486c32023-11-24T09:09:36ZengMDPI AGMathematics2227-73902022-11-011022432010.3390/math10224320Learning Adaptive Spatial Regularization and Temporal-Aware Correlation Filters for Visual Object TrackingLiqiang Liu0Tiantian Feng1Yanfang Fu2Chao Shen3Zhijuan Hu4Maoyuan Qin5Xiaojun Bai6Shifeng Zhao7School of Computer Science and Engineering, Xi’an Technological University, Xi’an 710021, ChinaSchool of Physics and Optoelectronic Engineering, Xidian University, Xi’an 710071, ChinaSchool of Computer Science and Engineering, Xi’an Technological University, Xi’an 710021, ChinaSchool of Computer Science and Engineering, Xi’an Technological University, Xi’an 710021, ChinaSchool of Computer Science and Engineering, Xi’an Technological University, Xi’an 710021, ChinaSchool of Computer Science and Engineering, Xi’an Technological University, Xi’an 710021, ChinaSchool of Computer Science and Engineering, Xi’an Technological University, Xi’an 710021, ChinaSchool of Computer Science and Engineering, Xi’an Technological University, Xi’an 710021, ChinaRecently, discriminative correlation filters (DCF) based trackers have gained much attention and obtained remarkable achievements for their high efficiency and outstanding performance. However, undesirable boundary effects occur when the DCF-based trackers suffer from challenging situations, such as occlusion, background clutters, fast motion, and so on. To address these problems, this work proposes a novel adaptive spatial regularization and temporal-aware correlation filters (ASTCF) model to deal with the boundary effects which occur in the correlation filters tracking. Firstly, our ASTCF model learns a more robust correlation filter template by introducing spatial regularization and temporal-aware components into the objective function. The adaptive spatial regularization provides a more robust appearance model to handle the large appearance changes at different times; meanwhile, the temporal-aware constraint can enhance the time continuity and consistency of this model. They make correlation filters model more discriminating, and also reduce the influence of the boundary effects during the tracking process. Secondly, the objective function can be transformed into three sub-problems with closed-form solutions and effectively solved via the alternating direction method of multipliers (ADMM). Finally, we compare our tracker with some representative methods and evaluate using three different benchmarks, including OTB2015, VOT2018 and LaSOT datasets, where the experimental results demonstrate the superiority of our tracker on most of the performance criteria compared with the existing trackers.https://www.mdpi.com/2227-7390/10/22/4320spatial regularizationtemporal-awarecorrelation filter trackingalternating direction method of multipliersboundary effect |
| spellingShingle | Liqiang Liu Tiantian Feng Yanfang Fu Chao Shen Zhijuan Hu Maoyuan Qin Xiaojun Bai Shifeng Zhao Learning Adaptive Spatial Regularization and Temporal-Aware Correlation Filters for Visual Object Tracking Mathematics spatial regularization temporal-aware correlation filter tracking alternating direction method of multipliers boundary effect |
| title | Learning Adaptive Spatial Regularization and Temporal-Aware Correlation Filters for Visual Object Tracking |
| title_full | Learning Adaptive Spatial Regularization and Temporal-Aware Correlation Filters for Visual Object Tracking |
| title_fullStr | Learning Adaptive Spatial Regularization and Temporal-Aware Correlation Filters for Visual Object Tracking |
| title_full_unstemmed | Learning Adaptive Spatial Regularization and Temporal-Aware Correlation Filters for Visual Object Tracking |
| title_short | Learning Adaptive Spatial Regularization and Temporal-Aware Correlation Filters for Visual Object Tracking |
| title_sort | learning adaptive spatial regularization and temporal aware correlation filters for visual object tracking |
| topic | spatial regularization temporal-aware correlation filter tracking alternating direction method of multipliers boundary effect |
| url | https://www.mdpi.com/2227-7390/10/22/4320 |
| work_keys_str_mv | AT liqiangliu learningadaptivespatialregularizationandtemporalawarecorrelationfiltersforvisualobjecttracking AT tiantianfeng learningadaptivespatialregularizationandtemporalawarecorrelationfiltersforvisualobjecttracking AT yanfangfu learningadaptivespatialregularizationandtemporalawarecorrelationfiltersforvisualobjecttracking AT chaoshen learningadaptivespatialregularizationandtemporalawarecorrelationfiltersforvisualobjecttracking AT zhijuanhu learningadaptivespatialregularizationandtemporalawarecorrelationfiltersforvisualobjecttracking AT maoyuanqin learningadaptivespatialregularizationandtemporalawarecorrelationfiltersforvisualobjecttracking AT xiaojunbai learningadaptivespatialregularizationandtemporalawarecorrelationfiltersforvisualobjecttracking AT shifengzhao learningadaptivespatialregularizationandtemporalawarecorrelationfiltersforvisualobjecttracking |