Abstract:The goal of example-based texture synthesis methods is to generate arbitrarily large textures from limited exemplars in order to fit the exact dimensions and resolution required for a specific modeling task. The challenge is to faithfully capture all of the visual characteristics of the exemplar texture, without introducing obvious repetitions or unnatural looking visual elements. While existing non-parametric synthesis methods have made remarkable progress towards this goal, most such methods have been demonstrated only on relatively low-resolution exemplars. Real-world high resolution textures often contain texture details at multiple scales, which these methods have difficulty reproducing faithfully. In this work, we present a new general-purpose and fully automatic self-tuning non-parametric texture synthesis method that extends Texture Optimization by introducing several key improvements that result in superior synthesis ability. Our method is able to self-tune its various parameters and weights and focuses on addressing three challenging aspects of texture synthesis: i irregular large scale structures are faithfully reproduced through the use of automatically generated and weighted guidance channels; ii repetition and smoothing of texture patches is avoided by new spatial uniformity constraints; iii a smart initialization strategy is used in order to improve the synthesis of regular and near-regular textures, without affecting textures that do not exhibit regularities. We demonstrate the versatility and robustness of our completely automatic approach on a variety of challenging high-resolution texture exemplars.

Non-Local Texture Optimization With Wasserstein Regularization Under Convolutional Neural Network

Non-stationary texture synthesis by adversarial expansion

Long Range Constraints for Neural Texture Synthesis Using Sliced Wasserstein Loss

Stationary Dynamic Texture Synthesis Using Convolutional Neural Networks

Conditional Generative ConvNets for Exemplar-Based Texture Synthesis

GAGCN: Generative Adversarial Graph Convolutional Network for Non-Homogeneous Texture Extension Synthesis

Self Tuning Texture Optimization

Exemplar Based Regular Texture Synthesis Using LSTM

Co-occurrence based texture synthesis

Texture Synthesis Through Convolutional Neural Networks and Spectrum Constraints

Combining Inconsistent Textures Using Convolutional Neural Networks.

Texture Networks: Feed-forward Synthesis of Textures and Stylized Images

Learning in a Single Domain for Non-Stationary Multi-Texture Synthesis

Generating Non-Stationary Textures using Self-Rectification

Texture Representation via Analysis and Synthesis with Generative Adversarial Networks

TextureWGAN: Texture Preserving WGAN with MLE Regularizer for Inverse Problems

Fast Texture Synthesis for Discrete Example-Based Elements

Edge-Based Video Compression Texture Synthesis Using Generative Adversarial Network

CNN based texture synthesize with Semantic segment

A Texture-Considerate Convolutional Neural Network Approach for Color Consistency in Remote Sensing Imagery

Paying U-Attention to Textures: Multi-Stage Hourglass Vision Transformer for Universal Texture Synthesis