一种用于视觉跟踪的低秩上下文感知的相关滤波器  

Correlation Filter Based on Low-rank and Context-aware for Visual Tracking

作　　者：苏银强 王宣[1] 王淳 李充 徐芳[1] SU Yinqiang;WANG Xuan;WANG Chun;LI Chong;XU Fang(Changchun Institute of Optics,Fine Mechanics and Physics(CIOMP),Chinese Academy of Sciences,Changchun 130000,China;University of Chinese Academy of Science,Beijing 100049,China;The First Military Representative Office of the Military Representative Bureau of the Army Equipment Department of the Chinese People’s Liberation Army in Changchun Shenyang,Changchun 130000,China)

机构地区：[1]中国科学院长春光学精密机械与物理研究所,长春130000 [2]中国科学院大学,北京100049 [3]中国人民解放军空军装备部驻沈阳地区军事代表局驻长春地区军事代表室,长春130000

出　　处：《计算机科学》2024年第9期121-128,共8页Computer Science

基　　金：国家自然科学基金面上项目(62175233);吉林省自然科学基金面上项目(20220101111JC)。

摘　　要：基于DCF的目标跟踪方法在保持实时运行时,由于在精度和鲁棒性之间实现了很好的权衡而备受关注。但是,当出现遮挡、移出视野、平面外旋转等干扰时,现有跟踪器仍面临着模型漂移甚至跟踪失败的情况。为此,提出了一种基于低秩上下文感知的相关滤波器LR_CACF。具体来说,在滤波器学习阶段,直接将目标及其上下文信息集成到DCF框架中,以更好地将目标从背景中鉴别出来;同时,对跨帧视频施加低秩约束以强调时序平滑性,使得学习的滤波器处于一个低维的鉴别流行上,进一步提高了跟踪性能;然后,利用ADMM实现滤波模型的高效优化;此外,针对模型失真的问题,启动多模态检测机制来识别响应图的可靠性,当反馈不可靠时,滤波器停止训练,同时扩大搜索区域,并采用区域重叠的方法重新捕获目标。在OTB-50,OTB-100和DTB70数据集上进行了大量实验,实验结果表明,相对于基线SAMF_CA,在DP方面,LR_CACF分别获得了6.9%,4.0%和7.1%的增益,AUC分别提高了3.6%,2.7%和5.4%。基于属性分析的结果表明,LR_CACF尤其擅长处理遮挡、移出视野、平面外旋转、低分辨率和快速运动等场景。Discriminative correlation filter(DCF)-based visual tracking approaches have attracted remarkable attention due to their good tradeoff between accuracy and robustness while running at real-time.However,the existing trackers still face model drift and even tracking failure situation when there are interferences such as long-term occlusion,out-of-view and out-of-plane rotation.To this end,we propose a low-rank and context-aware correlation filter(LR_CACF).Specifically,we directly integrate the target and its global contexts into DCF framework during filter learning stage to better discriminate the target from surrounding.Meanwhile,the low-rank constraint is injected across frames to emphasize the temporal smoothness,so that the learned filter is retained in a low-dimensional discriminant manifold to further improve tracking performance.Then,the ADMM is used to optimize the model effectively.Moreover,for model distortion,the multimodal detection mechanism is utilized to identify anomaly in the response.The filter stops training while extends the search regions to recapture the target when feedback is unreliable.Finally,extensive experiments are conducted on OTB50,OTB100 and DTB70 datasets,and the results demonstrate that,compared with the baseline SAMF_CA,LR_CACF achieves gains of 6.9%,4.0%and 7.1%in DP,respectively,and the average AUC improves by 3.6%,2.7%and 5.4%,respectively.Meanwhile,attribute-based evaluation shows that the proposed tracker is parti-cularly adept at handling the scenes such as occlusion,out-of-view,out-of-plane rotation,low resolution,and fast motion.

关 键 词：视觉跟踪 相关滤波 低秩约束 上下文感知 重检测 

分 类 号：TP394.1[自动化与计算机技术—计算机应用技术] TH691.9[自动化与计算机技术—计算机科学与技术]