OPT OpenIR  > 光学影像学习与分析中心
GAN and DCN based multi-step supervised learning for image semantic segmentation
Fang, Jie1,2; Cao, Xiaoqian3
2018
Conference Name1st Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2018
Source PublicationPattern Recognition and Computer Vision - First Chinese Conference, PRCV 2018, Proceedings
Volume11257 LNCS
Pages28-40
Conference Date2018-11-23
Conference PlaceGuangzhou, China
PublisherSpringer Verlag
Contribution Rank1
Abstract

Image semantic segmentation contains two sub-tasks, segmenting and labeling. However, the recent fully convolutional network (FCN) based methods often ignore the first sub-task and consider it as a direct labeling one. Even though these methods have achieved competitive performances, they obtained spatially fragmented and disconnected outputs. The reason is that, pixel-level relationships inside the deepest layers become inconsistent since traditional FCNs do not have any explicit pixel grouping mechanism. To address this problem, a multi-step supervised learning method, which contains image-level supervised learning step and pixel-level supervised learning step, is proposed. Specifically, as for the visualized result of image semantic segmentation, it is actually an image-to-image transformation problem, from RGB domain to category label domain. The recent conditional generative adversarial network (cGAN) has achieved significant performance for image-to-image generation task, and the generated image remains good regional connectivity. Therefore, a cGAN supervised by RGB-category label map is used to obtain a coarse segmentation mask, which avoids generating disconnected segmentation results to a certain extent. Furthermore, an interaction information (II) loss term is proposed for cGAN to remain the spatial structure of the segmentation mask. Additionally, dilated convolutional networks (DCNs) have achieved significant performance in object detection field, especially for small objects because of its special receptive field settings. Specific to image semantic segmentation, if each pixel is seen as an object, this task can be transformed to object detection. In this case, combined with the segmentation mask from cGAN, a DCN supervised by the pixel-level label is used to finalize the category recognition of each pixel in the image. The proposed method achieves satisfactory performances on three public and challenging datasets for image semantic segmentation. ? Springer Nature Switzerland AG 2018.

Department光学影像学习与分析中心
DOI10.1007/978-3-030-03335-4_3
Indexed ByEI
ISBN9783030033347
Language英语
ISSN03029743;16113349
EI Accession Number20184806152603
Citation statistics
Document Type会议论文
Identifierhttp://ir.opt.ac.cn/handle/181661/30869
Collection光学影像学习与分析中心
Corresponding AuthorFang, Jie
Affiliation1.Center for OPTical IMagery Analysis and Learning (OPTIMAL), Xi’an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi’an; Shaanxi; 710119, China;
2.University of Chinese Academy of Sciences, 19A Yuquanlu, Beijing; 100049, China;
3.College of Electrical and Information Engineering, Shaanxi University of Science and Technology, Xi’an; Shaanxi; 710021, China
Recommended Citation
GB/T 7714
Fang, Jie,Cao, Xiaoqian. GAN and DCN based multi-step supervised learning for image semantic segmentation[C]:Springer Verlag,2018:28-40.
Files in This Item:
File Name/Size DocType Version Access License
GAN and DCN based mu(997KB)会议论文 限制开放CC BY-NC-SAApplication Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Fang, Jie]'s Articles
[Cao, Xiaoqian]'s Articles
Baidu academic
Similar articles in Baidu academic
[Fang, Jie]'s Articles
[Cao, Xiaoqian]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Fang, Jie]'s Articles
[Cao, Xiaoqian]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.