SiamMCE: An Efficient Siamese Network for Real-Time Object Tracking With Dual-Correlation Strategy and Dynamic Template Updating
Real-time object tracking in dynamic environments poses significant challenges in balancing computational efficiency with robust performance under complex scenarios such as occlusion and illumination changes. This paper presents SiamMCE (Siamese MobileNet-CE) tracker, an optimized Siamese network va...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11079572/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839617284254990336 |
|---|---|
| author | Xin Wang Dejiang Wang Mingchao Sun Hang Ren Yulian Zhang Songwei Han Ligang Liu |
| author_facet | Xin Wang Dejiang Wang Mingchao Sun Hang Ren Yulian Zhang Songwei Han Ligang Liu |
| author_sort | Xin Wang |
| collection | DOAJ |
| description | Real-time object tracking in dynamic environments poses significant challenges in balancing computational efficiency with robust performance under complex scenarios such as occlusion and illumination changes. This paper presents SiamMCE (Siamese MobileNet-CE) tracker, an optimized Siamese network variant that integrates three key innovations to address these challenges. First, we design a lightweight backbone network MobileNet-CE which is based on MobileNetV3-Small through strategic integration of Convolutional Block Attention Module (CBAM) and Efficient Channel Attention Module (ECAM), reducing parameters by 18% while enhancing feature discriminability. Second, we propose a Dual-Correlation Strategy combining Pixel-wise Correlation (PWC) and Channel-wise Correlation (CWC) operations to improve localization precision through complementary spatial-channel feature fusion. Third, a Dynamic Template Adaptation mechanism leverages response map analysis via UpdateNet to enable online template refinement, mitigating drift accumulation during long-term tracking. Extensive experiments on benchmarks (OTB-2015, VOT2018, UAV123) demonstrate that SiamMCE achieves robust performance across mainstream tracking tasks, balancing competitive accuracy with real-time operation on embedded platforms. This capability enables new applications in dynamic environments, such as UAV-based detection and mobile surveillance, where sustained reliability is critical. |
| format | Article |
| id | doaj-art-474f13f7a8d34fa5b25f18f005c782b2 |
| institution | Matheson Library |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-474f13f7a8d34fa5b25f18f005c782b22025-07-24T23:01:51ZengIEEEIEEE Access2169-35362025-01-011312456912458610.1109/ACCESS.2025.358856811079572SiamMCE: An Efficient Siamese Network for Real-Time Object Tracking With Dual-Correlation Strategy and Dynamic Template UpdatingXin Wang0https://orcid.org/0009-0008-6181-8232Dejiang Wang1Mingchao Sun2Hang Ren3Yulian Zhang4Songwei Han5Ligang Liu6Chinese Academy of Sciences, Changchun Institute of Optics, Fine Mechanics and Physics, Changchun, ChinaChinese Academy of Sciences, Changchun Institute of Optics, Fine Mechanics and Physics, Changchun, ChinaChinese Academy of Sciences, Changchun Institute of Optics, Fine Mechanics and Physics, Changchun, ChinaChinese Academy of Sciences, Changchun Institute of Optics, Fine Mechanics and Physics, Changchun, ChinaState Key Laboratory of Dynamic Optical Imaging and Measurement, Changchun, ChinaChinese Academy of Sciences, Changchun Institute of Optics, Fine Mechanics and Physics, Changchun, ChinaChinese Academy of Sciences, Changchun Institute of Optics, Fine Mechanics and Physics, Changchun, ChinaReal-time object tracking in dynamic environments poses significant challenges in balancing computational efficiency with robust performance under complex scenarios such as occlusion and illumination changes. This paper presents SiamMCE (Siamese MobileNet-CE) tracker, an optimized Siamese network variant that integrates three key innovations to address these challenges. First, we design a lightweight backbone network MobileNet-CE which is based on MobileNetV3-Small through strategic integration of Convolutional Block Attention Module (CBAM) and Efficient Channel Attention Module (ECAM), reducing parameters by 18% while enhancing feature discriminability. Second, we propose a Dual-Correlation Strategy combining Pixel-wise Correlation (PWC) and Channel-wise Correlation (CWC) operations to improve localization precision through complementary spatial-channel feature fusion. Third, a Dynamic Template Adaptation mechanism leverages response map analysis via UpdateNet to enable online template refinement, mitigating drift accumulation during long-term tracking. Extensive experiments on benchmarks (OTB-2015, VOT2018, UAV123) demonstrate that SiamMCE achieves robust performance across mainstream tracking tasks, balancing competitive accuracy with real-time operation on embedded platforms. This capability enables new applications in dynamic environments, such as UAV-based detection and mobile surveillance, where sustained reliability is critical.https://ieeexplore.ieee.org/document/11079572/Siamese networkMobileNet-CEobject trackingdual-correlationtemplate updating |
| spellingShingle | Xin Wang Dejiang Wang Mingchao Sun Hang Ren Yulian Zhang Songwei Han Ligang Liu SiamMCE: An Efficient Siamese Network for Real-Time Object Tracking With Dual-Correlation Strategy and Dynamic Template Updating IEEE Access Siamese network MobileNet-CE object tracking dual-correlation template updating |
| title | SiamMCE: An Efficient Siamese Network for Real-Time Object Tracking With Dual-Correlation Strategy and Dynamic Template Updating |
| title_full | SiamMCE: An Efficient Siamese Network for Real-Time Object Tracking With Dual-Correlation Strategy and Dynamic Template Updating |
| title_fullStr | SiamMCE: An Efficient Siamese Network for Real-Time Object Tracking With Dual-Correlation Strategy and Dynamic Template Updating |
| title_full_unstemmed | SiamMCE: An Efficient Siamese Network for Real-Time Object Tracking With Dual-Correlation Strategy and Dynamic Template Updating |
| title_short | SiamMCE: An Efficient Siamese Network for Real-Time Object Tracking With Dual-Correlation Strategy and Dynamic Template Updating |
| title_sort | siammce an efficient siamese network for real time object tracking with dual correlation strategy and dynamic template updating |
| topic | Siamese network MobileNet-CE object tracking dual-correlation template updating |
| url | https://ieeexplore.ieee.org/document/11079572/ |
| work_keys_str_mv | AT xinwang siammceanefficientsiamesenetworkforrealtimeobjecttrackingwithdualcorrelationstrategyanddynamictemplateupdating AT dejiangwang siammceanefficientsiamesenetworkforrealtimeobjecttrackingwithdualcorrelationstrategyanddynamictemplateupdating AT mingchaosun siammceanefficientsiamesenetworkforrealtimeobjecttrackingwithdualcorrelationstrategyanddynamictemplateupdating AT hangren siammceanefficientsiamesenetworkforrealtimeobjecttrackingwithdualcorrelationstrategyanddynamictemplateupdating AT yulianzhang siammceanefficientsiamesenetworkforrealtimeobjecttrackingwithdualcorrelationstrategyanddynamictemplateupdating AT songweihan siammceanefficientsiamesenetworkforrealtimeobjecttrackingwithdualcorrelationstrategyanddynamictemplateupdating AT ligangliu siammceanefficientsiamesenetworkforrealtimeobjecttrackingwithdualcorrelationstrategyanddynamictemplateupdating |