标题:WEAKLY SUPERVISED IMAGE PARSING BY DISCRIMINATIVELY SEMANTIC GRAPH PROPAGATION
作者:Xu, Xiaocheng; Ma, Jun
作者机构:[Xu, Xiaocheng; Ma, Jun] Shandong Univ, Sch Comp Sci & Technol, Jinan, Peoples R China.
会议名称:IEEE International Conference on Multimedia & Expo (ICME)
会议日期:JUL 11-15, 2016
来源:2016 IEEE INTERNATIONAL CONFERENCE ON MULTIMEDIA & EXPO (ICME)
出版年:2016
卷:2016-August
DOI:10.1109/ICME.2016.7552866
关键词:Weakly supervised; image parsing; label co-occurrence; discriminatively; semantic graph propagation
摘要:In this paper, we concentrate on a challenging proble-mimage parsing trained on images with weakly supervised information, i.e.,image-level labels. Image-level labels are ambiguous and difficult for training. Typically, an affinity graph of superpixels is constructed to provide additional information about labels of the target superpixel. However, existing work constructs affinity graph in a naive manner, L1 reconstruction and k-NN are most used where label co-occurrence is a common phenomenon and degenerates the assignment performance. To overcome above problem, we proposed the use of discriminatively semantic ability between neighbor superpixels and the target superpixel in affinity graph construction. With simpler experiment setup and lower time complexity, our method achieves average per-class accuracy comparable to state-of-the-art performances in weakly-supervised image parsing task on datasets MSRC-21 and PASCAL VOC 2007.
收录类别:CPCI-S;EI;SCOPUS
资源类型:会议论文;期刊论文
原文链接:https://www.scopus.com/inward/record.uri?eid=2-s2.0-84987673755&doi=10.1109%2fICME.2016.7552866&partnerID=40&md5=bab93b93327c5612f4df080f25ef03d2
TOP