通过分析基于交并比(Intersection over union,IoU)预测的尺度估计模型的梯度更新过程,发现其在训练和推理过程仅将IoU作为度量,缺乏对预测框和真实目标框中心点距离的约束,导致外观模型更新过程中模板受到污染,前景和背景分类时定位出现偏差.基于此发现,构建了一种结合IoU和中心点距离的新度量NDIoU(Normalization distance IoU),在此基础上提出一种新的尺度估计方法,并将其嵌入判别式跟踪框架.即在训练阶段以NDIoU为标签,设计了具有中心点距离约束的损失函数监督网络的学习,在线推理期间通过最大化NDIoU微调目标尺度,以帮助外观模型更新时获得更加准确的样本.在七个数据集上与相关主流方法进行对比,所提方法的综合性能优于所有对比算法.特别是在GOT-10k数据集上,所提方法的AO、SR0.50和SR0.75三个指标达到了 65.4％、78.7％和53.4％,分别超过基线模型4.3％、7.0％和4.2％.
Accurate Scale Estimation With IoU and Distance Between Centroids for Object Tracking
This paper first analyzes the gradient update process of the scale estimation model of intersection over union(IoU)prediction in detail,and finds that when the IoU is used as a metric in the training and inference pro-cess,the target scale estimation in the tracking process is inaccurate due to the absence of the constraint on the dis-tance between the two centroids.As a result,the template is polluted in the updating process of the object appear-ance model,which cannot discriminate the target and environment.With this insight,we propose a new metric NDIoU(normalization distance IoU)that combines the IoU and distance between two centroids to estimate the tar-get scale and proposes a new scale estimation method,which is embedded into the discriminative tracking frame-work.Using NDIoU as the label to supervise the distance between centroids,it is incorporated into the loss func-tion to facilitate the learning of the network.During online inference,NDIoU is maximized to fine-tune the target scale.Finally,the proposed method is embedded into the discriminative tracking framework and compared with re-lated state-of-the-art methods on seven data sets.The extensive experiments demonstrate that our method outper-forms all the state-of-the-art algorithms.Especially,on the GOT-10k dataset,our method achieves 65.4％,78.7％and 53.4％on the three metrics of AO,SR0.50 and SR0.75,which are better than the baseline by 4.3％,7.0％and 4.2％,respectively.
Object trackingintersection over union(IoU)scale estimationdistance between centroids
国家科技期刊平台
NSTL
万方数据
