对象定位处理中分类信息融合技术研究

被引:4
作者
钱怡
林莹
武港山
机构
[1] 南京大学计算机软件新技术国家重点实验室
关键词
信息融合; 对象定位; 多类别分类; 多示例多标记学习框架; 快速子窗口搜索方法; 最优框;
D O I
暂无
中图分类号
TP391.41 []; TP202 [设计、性能分析与综合];
学科分类号
080203 ; 0811 ; 081101 ; 081102 ;
摘要
为提高图像中对象定位技术的处理效果,对对象定位技术和分类技术的融合进行了研究。针对大规模、多对象类别的图像对象定位问题,提出了先进行快速分类,再精确定位的处理方案。通过MIMLSVM+多类别分类算法预判出包含对象的图像,利用ESS方法在上述图像中定位对象;针对高精度对象定位需求,提出了融入全局分类信息的最优框打分机制,将MIMLSVM+算法对于图像的分类信息融入ESS方法中最优框的打分信息中。在PASCAL 2006数据集上相应的实验结果表明,前者在缩短处理时间的同时取得了不错的定位平均精度,而后者对最优框得分的改进也在多个类别上带来了定位效果的提高。实验结果表明,分类信息融入对象定位处理中能提升处理效果。
引用
收藏
页码:3844 / 3849
页数:6
相关论文
共 20 条
[1]   Distinctive image features from scale-invariant keypoints [J].
Lowe, DG .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2004, 60 (02) :91-110
[2]   A critical view of context [J].
Wolf, Lior ;
Bileschi, Stanley .
INTERNATIONAL JOURNAL OF COMPUTER VISION, 2006, 69 (02) :251-261
[3]  
MIMLRBF: RBF neural networks for multi-instance multi-label learning[J] . Min-Ling Zhang,Zhi-Jian Wang. &nbspNeurocomputing . 2009 (16)
[4]  
Multi-instance multi-label learning[J] . Zhi-Hua Zhou,Min-Ling Zhang,Sheng-Jun Huang,Yu-Feng Li. &nbspArtificial Intelligence . 2011 (1)
[5]  
M3MIML:A Maximum Margin Method for Multi-Instance Multi-Label Learning. Zhang Minling,Zhou Zhihua. Proc of the8th IEEE International Conference on Data Mining . 2008
[6]  
Beyond sliding windows: Object local-ization by efcient subwindow search. Lampert C H,Blaschko M B,Hofmann T. Proceedings of the2008IEEE Confer-ence on Computer Vision and Pattern Recognition . 2008
[7]  
A new SVM approach to multi-instance multi-label learning. NGUYEN N. Proc of the 10th IEEE International Conference on Data Mining . 2010
[8]  
Combining Efficient Object Localization and Image Classification. Harzallah, H,Jurie, F,Schmid, C. Proceedings of IEEE 12th International Conference on Computer Vision . 2009
[9]  
Drosophila gene expression pattern annotation through multi-instance multi-label learning. Li, Ying-Xin,Ji, Shuiwang,Kumar, Sudhir,Ye, Jieping,Zhou, Zhi-Hua. IEEE/ACM Transactions on Computational Biology and Bioinformatics . 2012
[10]  
Efficient subwindow search: a branch and bound framework for object localization. Lampert Christoph H,Blaschko Matthew B,Hofmann Thomas. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2009