Macrocanonical Models for Texture Synthesis

Abstract : In this article we consider macrocanonical models for texture synthesis. In these models samples are generated given an input texture image and a set of features which should be matched in expectation. It is known that if the images are quantized, macrocanonical models are given by Gibbs measures, using the maximum entropy principle. We study conditions under which this result extends to real-valued images. If these conditions hold, finding a macrocanonical model amounts to minimizing a convex function and sampling from an associated Gibbs measure. We analyze an algorithm which alternates between sampling and minimizing. We present experiments with neural network features and study the drawbacks and advantages of using this sampling scheme.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02093364
Contributor : Valentin de Bortoli <>
Submitted on : Monday, April 8, 2019 - 7:32:09 PM
Last modification on : Wednesday, May 1, 2019 - 1:18:32 AM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02093364, version 1

Citation

Valentin de Bortoli, Agnès Desolneux, Bruno Galerne, Arthur Leclaire. Macrocanonical Models for Texture Synthesis. 2019. ⟨hal-02093364⟩

Share

Metrics

Record views

15

Files downloads

101