TileGAN: Synthesis of Large-Scale Non-Homogeneous Textures

Abstract

We tackle the problem of texture synthesis in the setting where many input images are given and a large-scale output is required. We build on recent generative adversarial networks and propose two extensions in this paper. First, we propose an algorithm to combine outputs of GANs trained on a smaller resolution to produce a large-scale plausible texture map with virtually no boundary artifacts. Second, we propose a user interface to enable artistic control. Our quantitative and qualitative results showcase the generation of synthesized high-resolution maps consisting of up to hundreds of megapixels as a case in point.

arXiv commentCode is available at http://github.com/afruehstueck/tileGAN

References in corpus (11)

• Large Scale GAN Training for High Fidelity Natural Image Synthesis
• Self-Attention Generative Adversarial Networks
• Toward Multimodal Image-to-Image Translation
• Texture Synthesis Using Convolutional Neural Networks
• Non-Stationary Texture Synthesis by Adversarial Expansion
• Texture Synthesis with Spatial Generative Adversarial Networks
• CariGANs: Unpaired Photo-to-Caricature Translation
• FrankenGAN: Guided Detail Synthesis for Building Mass-Models Using Style-Synchonized GANs
• Multi-chart Generative Surface Modeling
• GANosaic: Mosaic Creation with Generative Texture Manifolds
• Copy the Old or Paint Anew? An Adversarial Framework for (non-) Parametric Image Stylization

Cited by in corpus (17)

• SEAN: Image Synthesis with Semantic Region-Adaptive Normalization
• Barbershop: GAN-based Image Compositing using Segmentation Masks
• Guided Fine-Tuning for Large-Scale Material Transfer
• Physics-informed GANs for Coastal Flood Visualization
• SeamlessGAN: Self-Supervised Synthesis of Tileable Texture Maps
• GramGAN: Deep 3D Texture Synthesis From 2D Exemplars
• NeuBTF: Neural fields for BTF encoding and transfer
• Transposer: Universal Texture Synthesis Using Feature Maps as Transposed Convolution Filter
• Compositional Neural Textures
• Neural Photometry-guided Visual Attribute Transfer
• Content-aware Tile Generation using Exterior Boundary Inpainting
• Directional Texture Editing for 3D Models
• Projective Urban Texturing
• Seamless Satellite-image Synthesis
• Deep Tiling: Texture Tile Synthesis Using a Deep Learning Approach
• Translate the Facial Regions You Like Using Region-Wise Normalization
• Neural-Polyptych: Content Controllable Painting Recreation for Diverse Genres