A novel hybrid augmented loss discriminator for text‐to‐image synthesis. Issue 2 (19th November 2020)

• Record Type: Journal Article
• Authors: Gan, Yan; Ye, Mao; Liu, Dan; Yang, Shangming; Xiang, Tao
• Is Part Of: International journal of intelligent systems. Volume 36:Issue 2(2021)
• Page Start: 1085
• Page End: 1107
• Publication Date: 2020-11-19
• Subjects: adversarial learning -- hybrid augmented loss -- perturbation -- text‐to‐image synthesis; Artificial intelligence -- Periodicals; Expert systems (Computer science) -- Periodicals; Intelligence artificielle -- Périodiques; Systèmes experts (Informatique) -- Périodiques; 006.3
• DOI: 10.1002/int.22333
• Languages: English
• ISSNs: 0884-8173

Abstract

Abstract: For the text‐to‐image synthesis task, most discriminators in existing generative adversarial networks based methods tend to fall into a local suboptimal state too early in the training process, resulting in the poor quality of generated images. To address the above problems, a hybrid augmented loss discriminator is designed. In this designed discriminator, to reduce the sensitivity of the discriminator classification recognition, make it pay attention to the semantic and structural changes, we add the loss value of the fake sample to the loss value of the real sample. Moreover, to indirectly guide the generator to generate samples, the loss value of the real sample is added to the fake sample. The loss value mixed with real and fake samples actually augments signal transmission. It perturbs parameter update of the discriminator during optimization and prevents the discriminator from falling into the local suboptimal state prematurely. Whereafter, we apply the proposed discriminator to two kinds of text‐to‐image synthesis tasks. Experimental results show that the proposed method can help the baseline models to improve performance.