High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANs

TL;DR

This paper introduces a high-resolution conditional GAN framework capable of generating photo-realistic images from semantic maps, with interactive editing features and improved diversity, surpassing existing methods in quality and resolution.

Contribution

The paper presents a novel high-resolution conditional GAN architecture with multi-scale generators and discriminators, enabling realistic image synthesis and interactive semantic manipulation.

Findings

Generated 2048x1024 images with high visual quality

Outperformed existing methods in human opinion studies

Enabled diverse and interactive image editing

Abstract

We present a new method for synthesizing high-resolution photo-realistic images from semantic label maps using conditional generative adversarial networks (conditional GANs). Conditional GANs have enabled a variety of applications, but the results are often limited to low-resolution and still far from realistic. In this work, we generate 2048x1024 visually appealing results with a novel adversarial loss, as well as new multi-scale generator and discriminator architectures. Furthermore, we extend our framework to interactive visual manipulation with two additional features. First, we incorporate object instance segmentation information, which enables object manipulations such as removing/adding objects and changing the object category. Second, we propose a method to generate diverse results given the same input, allowing users to edit the object appearance interactively. Human opinion studies demonstrate that our method significantly outperforms existing methods, advancing both the quality and the resolution of deep image synthesis and editing.