Open AccessFew Data Diversification in Training Generative Adversarial Networks
Author(s) -
Lucas Fontes Buzutti,
Carlos Eduardo Thomaz
Publication year - 2021
Language(s) - English
Resource type - Conference proceedings
DOI - 10.5753/wvc.2021.18892
Subject(s) - generative grammar , diversification (marketing strategy) , adversarial system , computer science , training set , image (mathematics) , artificial intelligence , training (meteorology) , sample (material) , variation (astronomy) , data modeling , image resolution , pattern recognition (psychology) , computer vision , geography , chemistry , physics , chromatography , marketing , meteorology , astrophysics , business , database
The first GANs have initially produced sharp images in relatively small resolution and with limited variations, and unstable training. Later works proposed new GAN models capable of generating sharp images in high resolution and with a high level of variation. However, these models use unlimited and highly diversified image sets. We discuss here the use of these models with real-world image sets, since they are composed of limited sample size sets.