Skip to main navigation Skip to search Skip to main content

LieMorph: Transformer-based Image Registration Using Flows on Lie Groups

Abstract

We propose a data-driven approach for deformable image registration that extends the TransMorph architecture with a module based on the flow equation formulated on matrix fields. Choosing a suitable Lie group allows us to precisely control components of the deformation that should be ignored by the regularization and provides a powerful plug-in replacement for the deformation field generator in learning-based methods. Combined with the TransMorph architecture, our method performs favorably on the IXI and OASIS datasets. Our implementation is available at https://github.com/sennhoj321/Liemorph.
Original languageEnglish
Number of pages12
Publication statusPublished - 2025
EventThe 36th British Machine Vision Conference (BMVC) - Cutler's Hall, Sheffield, Sheffield, United Kingdom
Duration: 24.11.202527.11.2025
https://bmvc2025.bmva.org

Conference

ConferenceThe 36th British Machine Vision Conference (BMVC)
Abbreviated titleBMVC 2025
Country/TerritoryUnited Kingdom
CitySheffield
Period24.11.2527.11.25
Internet address

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

  1. SDG 9 - Industry, Innovation, and Infrastructure
    SDG 9 Industry, Innovation, and Infrastructure

Fingerprint

Dive into the research topics of 'LieMorph: Transformer-based Image Registration Using Flows on Lie Groups'. Together they form a unique fingerprint.

Cite this