Zhou Wenzhang, Wen Longyin, Zhang Libo, Du Dawei, Luo Tiejian, Wu Yanjun
IEEE Trans Image Process. 2021;30:3597-3609. doi: 10.1109/TIP.2021.3060905. Epub 2021 Mar 17.
In this article, we present a novel Siamese center-aware network (SiamCAN) for visual tracking, which consists of the Siamese feature extraction subnetwork, followed by the classification, regression, and localization branches in parallel. The classification branch is used to distinguish the target from background, and the regression branch is introduced to regress the bounding box of the target. To reduce the impact of manually designed anchor boxes to adapt to different target motion patterns, we design the localization branch to localize the target center directly to assist the regression branch generating accurate results. Meanwhile, we introduce the global context module into the localization branch to capture long-range dependencies for more robustness to large displacements of the target. A multi-scale learnable attention module is used to guide these three branches to exploit discriminative features for better performance. Extensive experiments on 9 challenging benchmarks, namely VOT2016, VOT2018, VOT2019, OTB100, LTB35, LaSOT, TC128, UAV123 and VisDrone-SOT2019 demonstrate that SiamCAN achieves leading accuracy with high efficiency. Our source code is available at https://isrc.iscas.ac.cn/gitlab/research/siamcan.
在本文中，我们提出了一种用于视觉跟踪的新型暹罗中心感知网络（SiamCAN），它由暹罗特征提取子网络组成，随后并行地连接分类、回归和定位分支。分类分支用于区分目标与背景，回归分支用于回归目标的边界框。为了减少手动设计的锚框的影响以适应不同的目标运动模式，我们设计定位分支直接定位目标中心，以辅助回归分支生成准确结果。同时，我们将全局上下文模块引入定位分支，以捕捉远距离依赖关系，从而对目标的大位移具有更强的鲁棒性。使用多尺度可学习注意力模块来引导这三个分支利用判别性特征以获得更好的性能。在9个具有挑战性的基准测试上进行的大量实验，即VOT2016、VOT2018、VOT2019、OTB100、LTB35、LaSOT、TC128、UAV123和VisDrone-SOT2019，表明SiamCAN以高效率实现了领先的精度。我们的源代码可在https://isrc.iscas.ac.cn/gitlab/research/siamcan获取。
