Skip to main navigation Skip to search Skip to main content

A Geometrically Aware Auto-Encoder for Multi-texture Synthesis

  • Institut Polytechnique de Paris
  • Université Paris-Saclay

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

4 Citations (Scopus)

Abstract

We propose an auto-encoder architecture for multi-texture synthesis. The approach relies on both a compact encoder accounting for second order neural statistics and a generator incorporating adaptive periodic content. Images are embedded in a compact and geometrically consistent latent space, where the texture representation and its spatial organisation are disentangled. Texture synthesis and interpolation tasks can be performed directly from these latent codes. Our experiments demonstrate that our model outperforms state-of-the-art feed-forward methods in terms of visual quality and various texture related metrics. The code is available online.

Original languageEnglish
Title of host publicationScale Space and Variational Methods in Computer Vision - 9th International Conference, SSVM 2023, Proceedings
EditorsLuca Calatroni, Marco Donatelli, Serena Morigi, Marco Prato, Matteo Santacesaria
PublisherSpringer Science and Business Media Deutschland GmbH
Pages263-275
Number of pages13
ISBN (Print)9783031319747
DOIs
Publication statusPublished - 1 Jan 2023
Event9th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2023 - Santa Margherita di Pula, Italy
Duration: 21 May 202325 May 2023

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume14009 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference9th International Conference on Scale Space and Variational Methods in Computer Vision, SSVM 2023
Country/TerritoryItaly
CitySanta Margherita di Pula
Period21/05/2325/05/23

Keywords

  • auto-encoder
  • scale/orientation models
  • texture synthesis

Fingerprint

Dive into the research topics of 'A Geometrically Aware Auto-Encoder for Multi-texture Synthesis'. Together they form a unique fingerprint.

Cite this