Text-to-Image Synthesis: A Decade Survey

TL;DR

This survey reviews over 440 recent works on text-to-image synthesis, highlighting advances in models like GANs, autoregressive, and diffusion models, and discussing challenges, applications, and future directions in the field.

Contribution

It provides a comprehensive overview of recent developments in T2I, including model evolution, performance metrics, and emerging research challenges and opportunities.

Findings

GANs, autoregressive, and diffusion models are key to T2I development.

Advances have improved image quality, diversity, and controllability.

Current challenges include safety, personalization, and content consistency.

Abstract

When humans read a specific text, they often visualize the corresponding images, and we hope that computers can do the same. Text-to-image synthesis (T2I), which focuses on generating high-quality images from textual descriptions, has become a significant aspect of Artificial Intelligence Generated Content (AIGC) and a transformative direction in artificial intelligence research. Foundation models play a crucial role in T2I. In this survey, we review over 440 recent works on T2I. We start by briefly introducing how GANs, autoregressive models, and diffusion models have been used for image generation. Building on this foundation, we discuss the development of these models for T2I, focusing on their generative capabilities and diversity when conditioned on text. We also explore cutting-edge research on various aspects of T2I, including performance, controllability, personalized generation, safety concerns, and consistency in content and spatial relationships. Furthermore, we summarize the datasets and evaluation metrics commonly used in T2I research. Finally, we discuss the potential applications of T2I within AIGC, along with the challenges and future research opportunities in this field.