Citation

Tan, Yong Xuan and Lee, Chin Poo and Neo, Mai and Lim, Kian Ming and Lim, Jit Yan (2023) Enhanced Text-to-Image Synthesis With Self-Supervision. IEEE Access, 11. pp. 39508-39519. ISSN 2169-3536

Text 25.pdf - Published Version Restricted to Repository staff only Download (1MB)

Abstract

The task of Text-to-Image synthesis is a difficult challenge, especially when dealing with lowdata regimes, where the number of training samples is limited. In order to address this challenge, the SelfSupervision Text-to-Image Generative Adversarial Networks (SS-TiGAN) has been proposed. The method employs a bi-level architecture, which allows for the use of self-supervision to increase the number of training samples by generating rotation variants. This, in turn, maximizes the diversity of the model representation and enables the exploration of high-level object information for more detailed image construction. In addition to the use of self-supervision, SS-TiGAN also investigates various techniques to address the stability issues that arise in Generative Adversarial Networks. By implementing these techniques, the proposed SS-TiGAN has achieved a new state-of-the-art performance on two benchmark datasets, Oxford-102 and CUB. These results demonstrate the effectiveness of the SS-TiGAN method in synthesizing high-quality, realistic images from text descriptions under low-data regimes.

Item Type:	Article
Uncontrolled Keywords:	Text-to-image synthesis, generative model, GAN, self-supervised learning, generative adversarial networks.
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics T Technology > TA Engineering (General). Civil engineering (General) > TA1501-1820 Applied optics. Photonics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	02 Jun 2023 01:55
Last Modified:	31 Oct 2023 03:28
URII:	http://shdl.mmu.edu.my/id/eprint/11457

Downloads

Edit (login required)