Spectral normalized CycleGAN with application in semi-supervised semantic segmentation of sonar images

Zhisheng Zhang; Jingsong Tang; Peng Zhang; Mingqiang Ning; Han Li; Haoran Wu

doi:10.1117/12.2634721

4 May 2022 Spectral normalized CycleGAN with application in semi-supervised semantic segmentation of sonar images

Zhisheng Zhang, Jingsong Tang, Peng Zhang, Mingqiang Ning, Han Li, Haoran Wu

Proceedings Volume 12172, International Conference on Electronic Information Engineering and Computer Communication (EIECC 2021); 1217224 (2022) https://doi.org/10.1117/12.2634721
Event: International Conference on Electronic Information Engineering and Computer Communication (EIECC 2021), 2021, Nanchang, China

Abstract

The effectiveness of CycleGAN is demonstrated to outperform recent approaches for semi-supervised semantic segmentation on public segmentation benchmarks for a small number of the labelled data. However CycleGAN tends to generate same semantic segmentation results for acoustic image datasets, and can’t retain target details. To solve this problem, an spectral normalized CycleGAN network (SNCycleGAN) is presented, which applies spectral normalization to both generators and discriminators to stabilize the training of GANs. The experimental results demonstrate that semi-supervised training of SNCycleGAN helps to achieve reasonably accurate sonar targets segmentation from limited labelled data without using transfer learning, and surpass supervised training in detail preservation.

Citation Download Citation

Zhisheng Zhang, Jingsong Tang, Peng Zhang, Mingqiang Ning, Han Li, and Haoran Wu "Spectral normalized CycleGAN with application in semi-supervised semantic segmentation of sonar images", Proc. SPIE 12172, International Conference on Electronic Information Engineering and Computer Communication (EIECC 2021), 1217224 (4 May 2022); https://doi.org/10.1117/12.2634721

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available