Learning Local Structured Correlation Filters for Visual Tracking via Spatial Joint Regularization
Learning Local Structured Correlation Filters for Visual Tracking via Spatial Joint Regularization
复制标题
通过空间联合正则化学习用于视觉跟踪的局部结构化相关滤波器
DOI：
10.1109/access.2019.2906508
复制
发表时间：
2019-03
期刊：
影响因子：
3.9
通讯作者：
Huang Zhiqi
中科院分区：
文献类型：
--
作者：
Guo Chenggang;Chen Dongyi;Huang Zhiqi
鲁棒视觉跟踪是计算机视觉领域的一个基本问题，有着广泛的实际应用。近年来，鲁棒跟踪方法的研究主要集中在鉴别相关滤波器（DCF）上。然而，大多数基于DCF的方法开发他们的跟踪器的假设下，一个整体的外观模型，忽略了潜在的空间局部结构信息。在本文中，我们引入树结构的群稀疏正则化的DCF为基础的公式。将待学习的相关滤波器划分为分层局部组。通过在分层局部滤波器组的$l_{2}$ -范数上应用$l_{1}$ -范数来正则化响应和循环移位目标外观之间的关系。此外，局部响应一致性项与结构化稀疏度一起被并入，使得每个局部滤波器组对最终响应的贡献相等。采用加速邻近梯度法对该非光滑复合正则化问题进行优化。利用循环矩阵的性质，优化过程中的几个关键步骤可以在频域中有效地求解。四个公开的视觉跟踪基准进行实验。定量和定性评估表明，所提出的跟踪方法对一些国家的最先进的跟踪方法表现良好。
Robust visual tracking is a fundamental problem in the field of computer vision and has a wide range of practical applications. Recent progress in developing robust tracking methods are mainly made upon discriminative correlation filters (DCF). However, most DCF-based methods develop their trackers under the assumption of a holistic appearance model, ignoring the underlying spatial local structural information. In this paper, we introduce the tree-structured group sparsity regularization into the DCF-based formula. The correlation filter to be learned is divided into hierarchical local groups. The relationship between the response and the circularly shifted target appearance is regularized by applying the $l_{1}$ -norm across the $l_{2}$ -norm of the hierarchical local filter groups. Moreover, a local response consistency term is incorporated together with the structured sparsity to make each local filter group contributes equally to the final response. The accelerated proximal gradient method is employed to optimize this non-smooth composite regularization problem. Benefiting from the properties of circulant matrices, several key steps in the optimization process can be efficiently solved in the frequency domain. The experiments are conducted on four publicly available visual tracking benchmarks. Both quantitative and qualitative evaluations demonstrate that the proposed tracking method performs favorably against a number of state-of-the-art tracking methods.
登录
查看更多内容
相关滤波器学习视觉跟踪的峰值强度
影响因子：
11.8
作者：
Yao Sui;Guanghui Wang;Li Zhang
通讯作者：
Li Zhang
通过树结构组联合稀疏表示的多线索照明估计
影响因子：
19.5
作者：
Bing Li;Weihua Xiong;Weiming Hu （胡卫明）;Brian Funt;Junliang Xing
通讯作者：
Junliang Xing
DOI：
--
发表时间：
2010-09
期刊：
ArXiv
影响因子：
--
作者：
Jun Liu;Jieping Ye
通讯作者：
Jun Liu;Jieping Ye
DOI：
--
发表时间：
2009
期刊：
--
影响因子：
--
作者：
Xue Mei;Haibin Ling
通讯作者：
Xue Mei;Haibin Ling
影响因子：
3.9
作者：
Minkyu Lee;Taeoh Kim;Yuseok Ban;Eungyeol Song;Sangyoun Lee
通讯作者：
Minkyu Lee;Taeoh Kim;Yuseok Ban;Eungyeol Song;Sangyoun Lee