TextureGAN: Controlling Deep Image Synthesis with Texture Patches

TL;DR

TextureGAN introduces a novel method for deep image synthesis that allows users to control output textures by placing texture patches on sketches, enabling more realistic and user-guided image generation.

Contribution

The paper presents the first approach to incorporate explicit texture control in deep image synthesis using a local texture loss and a generative network.

Findings

The method produces plausible images faithful to user-specified textures.

Ablation studies demonstrate improved realism over existing methods.

The approach effectively synthesizes textures at arbitrary locations and scales.

Abstract

In this paper, we investigate deep image synthesis guided by sketch, color, and texture. Previous image synthesis methods can be controlled by sketch and color strokes but we are the first to examine texture control. We allow a user to place a texture patch on a sketch at arbitrary locations and scales to control the desired output texture. Our generative network learns to synthesize objects consistent with these texture suggestions. To achieve this, we develop a local texture loss in addition to adversarial and content loss to train the generative network. We conduct experiments using sketches generated from real images and textures sampled from a separate texture database and results show that our proposed algorithm is able to generate plausible images that are faithful to user controls. Ablation studies show that our proposed pipeline can generate more realistic images than adapting existing methods directly.