Answers are grounded in this paper and cite their sources.
A popular area of study in computer vision is image generation. How to improve the authenticity and diversity of generated images, and how to reduce the use of hardware resources and the size of model storage to make it more conducive to model landing are the focus and difficulty of image generation algorithms. Moreover, in practical applications, many other tasks are closely related to the image generation task, utilizing the created picture can lower the cost of data collecting, and the produced picture's quality directly influences how well some activities work. This article first introduces the definition of the generated model in supervised learning and unsupervised learning, and analyzes several reasons why learning should be generated into a model. It also introduces the development of the generated models in China and abroad, and makes a simple classification of the generated models with the previous depth degree.
No takes yet. Share an insight, caveat, or question.
Chang Liu (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: