Learning to Avoid Errors in GANs by Manipulating Input Spaces

TL;DR

This paper introduces a novel approach for reducing visual artifacts in GAN-generated images by manipulating input noise vectors, leading to improved image quality with minimal additional computational cost.

Contribution

It proposes a simple residual module that actively shifts input noise vectors away from error-prone regions, significantly reducing artifacts in GAN outputs.

Findings

Significantly fewer visual artifacts in generated images.

Minimal impact on diversity of outputs.

Low additional computational and memory costs.

Abstract

Despite recent advances, large scale visual artifacts are still a common occurrence in images generated by GANs. Previous work has focused on improving the generator's capability to accurately imitate the data distribution $p_{data}$. In this paper, we instead explore methods that enable GANs to actively avoid errors by manipulating the input space. The core idea is to apply small changes to each noise vector in order to shift them away from areas in the input space that tend to result in errors. We derive three different architectures from that idea. The main one of these consists of a simple residual module that leads to significantly less visual artifacts, while only slightly decreasing diversity. The module is trivial to add to existing GANs and costs almost zero computation and memory.