We propose a new approach for high resolution semantic image synthesis. It consists of one base image generator and multiple class-specific generators. The base generator generates high quality images based on a segmentation map. To further improve the quality of different objects, we create a bank of Generative Adversarial Networks (GANs) by separately training class-specific models. This has several benefits including -- dedicated weights for each class; centrally aligned data for each model; additional training data from other sources, potential of higher resolution and quality; and easy manipulation of a specific object in the scene. Experiments show that our approach can generate high quality images in high resolution while having flexibility of object-level control by using class-specific generators.
@article{arxiv.2110.04281,
title = {Collaging Class-specific GANs for Semantic Image Synthesis},
author = {Yuheng Li and Yijun Li and Jingwan Lu and Eli Shechtman and Yong Jae Lee and Krishna Kumar Singh},
journal= {arXiv preprint arXiv:2110.04281},
year = {2021}
}