arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.07560cs.CV
收藏

精确的无锚跟踪

Accurate Anchor Free Tracking

Tongji University(同济大学)

Shengyun Peng, Yunxuan Yu, Kun Wang, Lei He

摘要:

视觉目标跟踪是计算机视觉的一个重要应用。最近,基于暹罗的跟踪器已经实现了良好的准确性。然而,大多数基于暹罗的跟踪器效率不高,因为它们穷尽地搜索潜在的对象位置以定义锚,然后对每个锚进行分类(即,限界框)。本文提出了第一个无锚连体网络(AFSN)。具体地,目标对象由限界框中心、跟踪偏移和对象大小定义。所有这三个回归的暹罗网络没有额外的分类或区域的建议,并执行一次为每个帧。我们还调整了Siamese网络的步幅和感受野,并进一步进行消融实验,以定量地说明我们的AFSN的有效性。我们使用五个最常用的基准测试来评估AFSN,并将其与每个基准测试的最佳基于锚的跟踪器进行比较。AFSN比这些最好的基于锚的跟踪器快3-425倍。AFSN在基准集OTB 2015、VOT 2015、VOT 2016、VOT 2018和TrackingNet的所有指标方面的准确度也高出5.97%至12.4%,除了SiamRPN++在VOT 2018上的预期平均重叠(EAO)方面比AFSN高出4%(但SiamRPN++慢3.9倍)。

英文摘要:

Visual object tracking is an important application of computer vision. Recently, Siamese based trackers have achieved good accuracy. However, most of Siamese based trackers are not efficient, as they exhaustively search potential object locations to define anchors and then classify each anchor (i.e., a bounding box). This paper develops the first Anchor Free Siamese Network (AFSN). Specifically, a target object is defined by a bounding box center, tracking offset, and object size. All three are regressed by Siamese network with no additional classification or regional proposal, and performed once for each frame. We also tune the stride and receptive field for Siamese network, and further perform ablation experiments to quantitatively illustrate the effectiveness of our AFSN. We evaluate AFSN using five most commonly used benchmarks and compare to the best anchor-based trackers with source codes available for each benchmark. AFSN is 3-425 times faster than these best anchor based trackers. AFSN is also 5.97% to 12.4% more accurate in terms of all metrics for benchmark sets OTB2015, VOT2015, VOT2016, VOT2018 and TrackingNet, except that SiamRPN++ is 4% better than AFSN in terms of Expected Average Overlap (EAO) on VOT2018 (but SiamRPN++ is 3.9 times slower).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑