Feature-Grounded Single-Stage Text-to-Image Generation

作     者：Yuan Zhou Peng Wang Lei Xiang Haofeng Zhang 

作者机构：School of Artificial IntelligenceNanjing University of Information Science and TechnologyNanjing 210044China School of Computer Science and EngineeringNanjing University of Science and TechnologyNanjing 210094China 

出 版 物：《Tsinghua Science and Technology》 (清华大学学报（自然科学版（英文版）)

年 卷 期：2024年第29卷第2期

页      面：469-480页

核心收录：

学科分类：08[工学] 081104[工学-模式识别与智能系统] 0811[工学-控制科学与工程] 081201[工学-计算机系统结构] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported by the National Natural Science Foundation of China(No.61872187). 

主　　题：text-to-image(T2I) feature-grounded single-stage generation Generative Adversarial Network(GAN) 

摘      要：Recently,Generative Adversarial Networks(GANs)have become the mainstream text-to-image(T2I)framework.However,a standard normal distribution noise of inputs cannot provide sufficient information to synthesize an image that approaches the ground-truth image distribution.Moreover,the multistage generation strategy results in complex T2I applications.Therefore,this study proposes a novel feature-grounded single-stage T2I model,which considers the“realdistribution learned from training images as one input and introduces a worst-case-optimized similarity measure into the loss function to enhance the model s generation capacity.Experimental results on two benchmark datasets demonstrate the competitive performance of the proposed model in terms of the Frechet inception distance and inception score compared to those of some classical and state-of-the-art models,showing the improved similarities among the generated image,text,and ground truth.