? 基于卷积特征处理光照和遮挡的鲁棒跟踪算法
Journal of Computer Science and Technology
Quick Search in JCST
 Advanced Search 
      Home | PrePrint | SiteMap | Contact Us | Help
 
Indexed by   SCIE, EI ...
Bimonthly    Since 1986
Journal of Computer Science and Technology 2018, Vol. 33 Issue (1) :223-236    DOI: 10.1007/s11390-017-1764-5
Regular Paper << Previous Articles | >>
基于卷积特征处理光照和遮挡的鲁棒跟踪算法
Kang Li1,2,3, Fa-Zhi He1,2,*, Senior Member, CCF, Hai-Ping Yu1,2
1 State Key Laboratory of Software Engineering, Wuhan University, Wuhan 430072, China;
2 School of Computer Science, Wuhan University, Wuhan 430072, China;
3 School of Computer Science and Information Engineering, Hubei University, Wuhan 430062, China
Robust Visual Tracking Based on Convolutional Features with Illumination and Occlusion Handing
Kang Li1,2,3, Fa-Zhi He1,2,*, Senior Member, CCF, Hai-Ping Yu1,2
1 State Key Laboratory of Software Engineering, Wuhan University, Wuhan 430072, China;
2 School of Computer Science, Wuhan University, Wuhan 430072, China;
3 School of Computer Science and Information Engineering, Hubei University, Wuhan 430062, China

摘要
参考文献
相关文章
Download: [PDF 7139KB]  
摘要 视觉跟踪是计算机视觉中的一个重要领域。如何处理照明和遮挡问题是一个具有挑战性的问题。本文提出了一种新的和有效的跟踪算法来处理该问题。一方面,目标的初始外观总是具有清晰轮廓,具有光照不变性和鲁棒性。另一方面,特征在跟踪中起着重要作用,其中卷积特征具备良好性能。因此,我们采用卷积轮廓特征来描述目标外观。理论上,通过一阶边缘梯度算子对图像进行卷积来检测轮廓是有效的。特别是,Prewitt算子对水平边缘和垂直方向边缘更敏感,而Sobel算子对对角边缘更敏感,这样Prewitt算子和索贝尔具有内在互补性互补。技术上讲,设计了两组Prewitt和Sobel边缘检测算子来提取出一套完整的卷积特征,包括水平、垂直和对角边缘特征。在第一帧中,通过提取目标的轮廓特征来构造初始外观模型。通过分析这些轮廓特征的实验图像,可以发现,明亮的部分往往提供更多有用的信息来描述目标特性。因此,我们提出一种使用明亮的像素方法来比较候选样本和训练模型之间的相似性,从而可以处理部分遮挡问题。在得到新目标之后,提出了适应外观变化的、增量化的在线跟新策略。实验结果表明,通过集成Prewitt和Sobel边缘检测算作所提取的卷积特征可以有效学习鲁棒的外观模型。在9个具有挑战性的视频序列上的一系列实验结果表明,所提方法相比目前方法是非常有效和鲁棒。
关键词视频跟踪   卷积特征   梯度算子   在线学习   粒子滤波     
Abstract: Visual tracking is an important area in computer vision. How to deal with illumination and occlusion problems is a challenging issue. This paper presents a novel and efficient tracking algorithm to handle such problems. On one hand, a target's initial appearance always has clear contour, which is light-invariant and robust to illumination change. On the other hand, features play an important role in tracking, among which convolutional features have shown favorable performance. Therefore, we adopt convolved contour features to represent the target appearance. Generally speaking, first-order derivative edge gradient operators are efficient in detecting contours by convolving them with images. Especially, the Prewitt operator is more sensitive to horizontal and vertical edges, while the Sobel operator is more sensitive to diagonal edges. Inherently, Prewitt and Sobel are complementary with each other. Technically speaking, this paper designs two groups of Prewitt and Sobel edge detectors to extract a set of complete convolutional features, which include horizontal, vertical and diagonal edges features. In the first frame, contour features are extracted from the target to construct the initial appearance model. After the analysis of experimental image with these contour features, it can be found that the bright parts often provide more useful information to describe target characteristics. Therefore, we propose a method to compare the similarity between candidate sample and our trained model only using bright pixels, which makes our tracker able to deal with partial occlusion problem. After getting the new target, in order to adapt appearance change, we propose a corresponding online strategy to incrementally update our model. Experiments show that convolutional features extracted by well-integrated Prewitt and Sobel edge detectors can be efficient enough to learn robust appearance model. Numerous experimental results on nine challenging sequences show that our proposed approach is very effective and robust in comparison with the state-of-the-art trackers.
Keywordsvisual tracking   convolutional feature   gradient operator   online learning   particle filter     
Received 2016-11-15;
本文基金:

This paper is supported by the National Natural Science Foundation of China under Grant No. 61472289 and the National Key Research and Development Project of China under Grant No. 2016YFC0106305.

通讯作者: Fa-Zhi He     Email: fzhe@whu.edu.cn
About author: Kang Li is currently an assistant professor of School of Computer and Information Engineering of Hubei University, Wuhan. He received his B.S. degree in management from Anhui University, Hefei, in 2008. He received his M.S. degree in computer science from Huazhong Normal University, Wuhan, in 2012, and his Ph.D. degree in computer science from Wuhan University, Wuhan, in 2016. His research interests are computer vision, pattern recognition, image processing and computer graphics.
引用本文:   
Kang Li, Fa-Zhi He, Hai-Ping Yu.基于卷积特征处理光照和遮挡的鲁棒跟踪算法[J]  Journal of Computer Science and Technology , 2018,V33(1): 223-236
Kang Li, Fa-Zhi He, Hai-Ping Yu.Robust Visual Tracking Based on Convolutional Features with Illumination and Occlusion Handing[J]  Journal of Computer Science and Technology, 2018,V33(1): 223-236
链接本文:  
http://jcst.ict.ac.cn:8080/jcst/CN/10.1007/s11390-017-1764-5
Copyright 2010 by Journal of Computer Science and Technology