Scene Understanding Based on High-Order Potentials and Generative Adversarial Networks
Joint Authors
Zhao, Xiaoli
Wang, Guozhong
Zhang, Jiaqi
Zhang, Xiang
Source
Issue
Vol. 2018, Issue 2018 (31 Dec. 2018), pp.1-8, 8 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2018-08-05
Country of Publication
Egypt
No. of Pages
8
Main Subjects
Information Technology and Computer Science
Abstract EN
Scene understanding is to predict a class label at each pixel of an image.
In this study, we propose a semantic segmentation framework based on classic generative adversarial nets (GAN) to train a fully convolutional semantic segmentation model along with an adversarial network.
To improve the consistency of the segmented image, the high-order potentials, instead of unary or pairwise potentials, are adopted.
We realize the high-order potentials by substituting adversarial network for CRF model, which can continuously improve the consistency and details of the segmented semantic image until it cannot discriminate the segmented result from the ground truth.
A number of experiments are conducted on PASCAL VOC 2012 and Cityscapes datasets, and the quantitative and qualitative assessments have shown the effectiveness of our proposed approach.
American Psychological Association (APA)
Zhao, Xiaoli& Wang, Guozhong& Zhang, Jiaqi& Zhang, Xiang. 2018. Scene Understanding Based on High-Order Potentials and Generative Adversarial Networks. Advances in Multimedia،Vol. 2018, no. 2018, pp.1-8.
https://search.emarefa.net/detail/BIM-1118476
Modern Language Association (MLA)
Zhao, Xiaoli…[et al.]. Scene Understanding Based on High-Order Potentials and Generative Adversarial Networks. Advances in Multimedia No. 2018 (2018), pp.1-8.
https://search.emarefa.net/detail/BIM-1118476
American Medical Association (AMA)
Zhao, Xiaoli& Wang, Guozhong& Zhang, Jiaqi& Zhang, Xiang. Scene Understanding Based on High-Order Potentials and Generative Adversarial Networks. Advances in Multimedia. 2018. Vol. 2018, no. 2018, pp.1-8.
https://search.emarefa.net/detail/BIM-1118476
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-1118476