计算机工程与应用2024，Vol.60Issue(8)：213-224,12.DOI:10.3778/j.issn.1002-8331.2211-0435
分布统计特征的孪生网络目标跟踪方法
Siamese Networks for Object Tracking on Statistical Characteristics of Distributions
摘要
Abstract
Although siamese trackers have achieved great success,the tracking performance is inferior in complex scenes such as ambiguous boundaries.Most of the existing methods use the inflexible Dirac distribution for target localization.Due to the lack of uncertainty estimation of the bounding box,the target cannot be accurately located under the ambiguous boundaries.For this purpose,this paper improves SiamBAN.Firstly,the representation of the bounding box is changed from Dirac distribution to the general distribution within a certain range with the help of the characteristics that distribu-tion statistics of the bounding box are highly correlated with the actual localization quality.Secondly,a higher localization quality estimation score is generated by putting the distribution statistics into distribution guided quality predictor.Finally,classification and localization quality estimation are represented jointly which can overcome the problem of inconsistency between classification and localization in training and testing stages.Extensive experiments on visual tracking datasets including VOT2018,VOT2019,OTB100,UAV123,LaSOT,TrackingNet,and GOT-10k demonstrate that the performance of proposed method surpass SiamBAN by 3.3%～10%in terms of accuracy and EAO.关键词
孪生网络/定位质量/不确定性估计/分布统计特性/分布引导质量预测器Key words
siamese network/localization quality/uncertainty estimation/distribution statistics/distribution guided quality predictor分类
信息技术与安全科学引用本文复制引用
李俊,曹林,张帆,杜康宁,郭亚男..分布统计特征的孪生网络目标跟踪方法[J].计算机工程与应用,2024,60(8):213-224,12.基金项目
国家自然科学基金(62001033,62201066,U20A20163) （62001033,62201066,U20A20163）
北京信息科技大学"勤信人才"培育计划(QXTCP C202108) （QXTCP C202108）
北京市教委科研计划(KM202011232021,KZ202111232049,KM202111232014). （KM202011232021,KZ202111232049,KM202111232014）
