Multispectral Style Distances and Application to Texture Synthesis Using RGB Convolutional Neural Networks

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

Abstract

State-of-the-art methods for RGB texture synthesis and style transfer leverage the representations learned by convolutional neural networks (CNN) on large datasets. Style distances, obtained by comparing statistics of deep features, play a pivotal role in synthesis procedures. Extending these distances to multispectral images is challenging because the pre-trained CNN only operate on RGB images. This work presents two multispectral style distances that still rely on a RGB CNN to avoid additional training. The first consists in a classical style distance, averaged over images formed by triplets of spectral bands. The second takes advantage of a projection of the multispectral pixels onto a three-dimensional space. We demonstrate their efficiency by performing multispectral texture synthesis.

Original languageEnglish
Title of host publication2024 14th Workshop on Hyperspectral Imaging and Signal Processing
Subtitle of host publicationEvolution in Remote Sensing, WHISPERS 2024
PublisherIEEE Computer Society
ISBN (Electronic)9798331513139
DOIs
Publication statusPublished - 1 Jan 2024
Event14th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing, WHISPERS 2024 - Helsinki, Finland
Duration: 9 Dec 202411 Dec 2024

Publication series

NameWorkshop on Hyperspectral Image and Signal Processing, Evolution in Remote Sensing
ISSN (Print)2158-6276

Conference

Conference14th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing, WHISPERS 2024
Country/TerritoryFinland
CityHelsinki
Period9/12/2411/12/24

Keywords

  • Style distance
  • convolutional neural networks
  • deep learning
  • multispectral imaging
  • texture synthesis

Fingerprint

Dive into the research topics of 'Multispectral Style Distances and Application to Texture Synthesis Using RGB Convolutional Neural Networks'. Together they form a unique fingerprint.

Cite this