李冬梅,李涛,向涛

• 1:河南广播电视大学信息工程学院
• 2:电子科技大学计算机科学与工程学院

摘要(Abstract)：

针对复杂场景中车辆由于视角变化引起的检测精确度过低的问题,改进霍夫投票目标检测模型,提出一种在统一框架下通过不同权重组合发现目标最优视角并进行精确定位的方法。首先,利用一种无监督方法实现多视角车辆的子视角划分;其次,利用子视角划分结果定义霍夫投票过程中各正例样本在不同视角下的投票权重;最后,利用子视角划分和投票权重,提出一种新的适用于多视角目标检测的加权霍夫投票模型。在MITStreet Scene Cars和PASCAL VOC2007 Cars两个常用数据集上的实验结果表明,所提方法在不增加模型复杂度的前提下,有效提升了多视角目标检测精确度。

关键词(KeyWords)： 复杂场景;霍夫投票;最优视角;多视角目标检测;子类划分;局部线性嵌入(LLE)

Abstract:

Keywords:

基金项目(Foundation): 河南省科技厅科技攻关项目(182102210574);; 河南省教育厅重大专项项目(17A520065)~~

作者(Author): 李冬梅,李涛,向涛

DOI: 10.16652/j.issn.1004-373x.2018.15.017

参考文献(References)：

• [1]DALAL N,TRIGGS B.Histograms of oriented gradients for human detection[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition.San Diego:IEEE,2005:886-893.
• [2]WANG X,HAN T X,YAN S C.An HOG-LBP human detector with partial occlusion handling[C]//2009 IEEE International Conference on Computer Vision.Kyoto:IEEE,2009:32-39.
• [3]FELZENSZWALB P F,GIRSHICK R B,MCALLESTER D,et al.Object detection with discriminatively trained part-based models[J].IEEE transactions on pattern analysis and machine intelligence,2010,32(9):1627-1645.
• [4]PIRSIAVASH H,RAMANAN D.Steerable part models[C]//2012 IEEE Conference on Computer Vision and Pattern Recognition.Providence:IEEE,2012:3226-3233.
• [5]SEEMANN E,LEIBE B,SCHIELE B.Human multi-aspect detection of articulated objects[C]//2006 IEEE Conference on Computer Vision and Pattern Recognition.New York:IEEE,2006:1582-1588.
• [6]KUO C H,NEVATIA R.Robust multi-view car detection using unsupervised sub-categorization[C]//2009 IEEE Workshop on Applications of Computer Vision Workshop.Snowbird:IEEE,2009:1-8.
• [7]SHAN Y,HAN F,SAWHNEY H,et al.Learning exemplarbased categorization for the detection of multiview multi-pose objects[C]//2006 IEEE Conference on Computer Vision and Pattern Recognition.New York:IEEE,2006:1431-1438.
• [8]WU Bo,NEVATIA R.Cluster boosted tree classifier for multiview,multi-pose object detection[C]//2007 IEEE International Conference on Computer Vision.Rio de Janeiro:IEEE,2007:1-8.
• [9]KUSHAL A,SCHMID C,PONCE J.Flexible object models for category-level 3D object recognition[C]//2007 IEEE Conference on Computer Vision and Pattern Recognition.Minneapolis:IEEE,2007:1-8.
• [10]SHRIVASTAVA A,GUPTA A.Building part-based object detectors via 3D geometry[C]//2013 IEEE International Conference on Computer Vision.Sydney:IEEE,2013:1745-1752.
• [11]GLASNER D,GALUN M,ALPERT S,et al.Viewpointaware object detection and pose estimation[C]//2011 IEEE International Conference on Computer Vision and Pattern Recognition.Barcelona:IEEE,2011:1745-1753.
• [12]HAO S,MIN S,LI F F,et al.Learning a dense multi-view representation for detection,viewpoint classification and synthesis of object categories[C]//2009 IEEE International Conference on Computer Vision and Pattern Recognition.Kyoto:IEEE,2009:213-220.
• [13]LEIBE B,LEONARDIS A,SCHIELE B.Robust object detection with interleaved categorization and segmentation[J].International journal of computer vision,2008,77(1/3):259-289.
• [14]BILESCHI S M.Street Scenes:towards scene understanding in still images[D].Cambridge:Massachusetts Institute of Technology,Department of Electrical Engineering and Computer Science,2006.
• [15]EVERINGHAM M,VAN GOOL L,WILLIAMS C K I,et al.The PASCAL visual object classes challenge[DB/OL].[2007-02-12].http://host.robots.ox.ac.uk:8080/pascal/VOC/voc2007/index.html,2007.
• [16]GALL J,LEMPITSKY V.Class-specific Hough forests for object detection[C]//2009 IEEE International Conference on Computer Vision and Pattern Recognition.Miami:IEEE,2009:1022-1029.
• [17]APPEL R,PERONA P,BELONGIE S.Fast feature pyramids for object detection[J].IEEE transactions on pattern analysis and machine intelligence,2014,36(8):1532-1545.
• [18]向涛,李涛,赵雪专,等.基于随机森林的精确目标检测方法[J].计算机应用研究,2016,33(9):2837-2840.XIANG T,LI T,ZHAO X Z,et al.Random forests for accurate object detection[J].Application research of computers,2016,33(9):2837-2840.
• [19]LI T,YE M,DING J.Discriminative Hough context model for object[J].Visual computer,2014,33(11):59-69.