DeepAI AI Chat
Log In Sign Up

Texture Mixing by Interpolating Deep Statistics via Gaussian Models

by   Zi-Ming Wang, et al.

Recently, enthusiastic studies have devoted to texture synthesis using deep neural networks, because these networks excel at handling complex patterns in images. In these models, second-order statistics, such as Gram matrix, are used to describe textures. Despite the fact that these model have achieved promising results, the structure of their parametric space is still unclear, consequently, it is difficult to use them to mix textures. This paper addresses the texture mixing problem by using a Gaussian scheme to interpolate deep statistics computed from deep neural networks. More precisely, we first reveal that the statistics used in existing deep models can be unified using a stationary Gaussian scheme. We then present a novel algorithm to mix these statistics by interpolating between Gaussian models using optimal transport. We further apply our scheme to Neural Style Transfer, where we can create mixed styles. The experiments demonstrate that our method can achieve state-of-the-art results. Because all the computations are implemented in closed forms, our mixing algorithm adds only negligible time to the original texture synthesis procedure.


page 9

page 10

page 11

page 12

page 13

page 14

page 15


Optimal Textures: Fast and Robust Texture Synthesis and Style Transfer through Optimal Transport

This paper presents a light-weight, high-quality texture synthesis algor...

Conditional Generative ConvNets for Exemplar-based Texture Synthesis

The goal of exemplar-based texture synthesis is to generate texture imag...

Wasserstein Style Transfer

We propose Gaussian optimal transport for Image style transfer in an Enc...

Investigating the hybrid textures of neutrino mass matrix for near maximal atmospheric neutrino mixing

In the present paper, we have studied that the implication of a large va...

Texture Interpolation for Probing Visual Perception

Texture synthesis models are important to understand visual processing. ...

Pitfalls of the Gram Loss for Neural Texture Synthesis in Light of Deep Feature Histograms

Neural texture synthesis and style transfer are both powered by the Gram...

Long Range Constraints for Neural Texture Synthesis Using Sliced Wasserstein Loss

In the past decade, exemplar-based texture synthesis algorithms have see...