Label localization with weakly spatial constrained graph propagation

doi:10.1109/ICME.2013.6607502

	Label localization with weakly spatial constrained graph propagation
	Yu, Lei 1; Liu, Jing 1; Xu, Changsheng 1; Zhou, Xi 2
	2013
会议日期	July 15, 2013 - July 19, 2013
会议地点	San Jose, CA, United states
DOI	10.1109/ICME.2013.6607502
英文摘要	Properly utilizing the spatial correlation of regions benefits for improving the performance of label localization task. However, we could not obtain this information directly since we do not have the region level ground truth. In this paper, we propose a weakly spatial constrained graph propagation by mining the spatial correlation from unlabeled regions and integrating it into the graph propagation framework. Our main framework contains two steps: the spatial constrained graph (SCG) construction and label propagation. Firstly, images are over-segmented and each patch is considered as a node. We deem the relatively stable patch combination as a spatial context to construct the SCG, and encourage label propagations where those patches are visually similar as well as spatially consistent. In the second step, we add the dissimilarity constraints and image level label constraints to the label propagation. The propagation procedure is formulated as a constrained optimization problem and it can be efficiently solved by an iteration method. Experiments on three benchmark datasets demonstrate that the spatial correlation mined by our method is effective to the label localization task. © 2013 IEEE.
会议录	2013 IEEE International Conference on Multimedia and Expo, ICME 2013
语种	英语
电子版国际标准刊号	1945788X
ISSN号	19457871
内容类型	会议论文
源URL	[http://119.78.100.138/handle/2HOD01W0/4777]
专题	中国科学院重庆绿色智能技术研究院
作者单位	1.Institute of Automation, Chinese Academy of Sciences, 95 Zhongguancun East Road, 100190 Beijing, China; 2.Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, 85 Jinyu Avenue, New North Zone, 401122, China
推荐引用方式 GB/T 7714	Yu, Lei,Liu, Jing,Xu, Changsheng,et al. Label localization with weakly spatial constrained graph propagation[C]. 见:. San Jose, CA, United states. July 15, 2013 - July 19, 2013.