Flow Matching-enabled Test-Time Refinement for Unsupervised Cardiac MR Registration

Yunguan Fu, Wenjia Bai, Wen Yan, Matthew J Clarkson, Rhodri Huw Davies +1 more
3/1/2026
cs.CV

Official code: mathpluscode/FlowReg (2 stars, MIT, last commit 2026-03-09).

Abstract

Diffusion-based unsupervised image registration has been explored for cardiac cine MR, but expensive multi-step inference limits practical use. We propose FlowReg, a flow-matching framework in displacement field space that achieves strong registration in as few as two steps and supports further refinement with more steps. FlowReg uses warmup-reflow training: a single-step network first acts as a teacher, then a student learns to refine from arbitrary intermediate states, removing the need for a pre-trained model as in existing methods. An Initial Guess strategy feeds back the model prediction as the next starting point, improving refinement from step two onward. On ACDC and MM2 across six tasks (including cross-dataset generalization), FlowReg outperforms the state of the art on five tasks (+0.6% mean Dice score on average), with the largest gain in the left ventricle (+1.09%), and reduces LVEF estimation error on all six tasks (-2.58 percentage points), using only 0.7% extra parameters and no segmentation labels. Code is available at https://github.com/mathpluscode/FlowReg.

View on arXivView PDF

Code Implementations(1)

mathpluscode/FlowRegOfficial100%
MIT

Flow Matching-enabled Test-Time Refinement for Unsupervised Cardiac MR Registration

21PythonFeb 26, 20267 months agoMIT
cardiaccmrflowmatchingregistration

Similar Papers

Deep-learning-based groupwise registration for motion correction of cardiac T₁ mapping

82% match2024

A Deep Learning Based Method for Fast Registration of Cardiac Magnetic Resonance Images

81% match2025

MemWarp: Discontinuity-Preserving Cardiac Registration with Memorized Anatomical Filters

81% match1 repo72024

Highly efficient non-rigid registration in k-space with application to cardiac Magnetic Resonance Imaging

80% match1 repo2024

TLRN: Temporal Latent Residual Networks For Large Deformation Image Registration

80% match1 repo62024

FlowECG: Using Flow Matching to Create a More Efficient ECG Signal Generator

80% match2025

Cite this paper

@article{fu2026flow,
  title  = {Flow Matching-enabled Test-Time Refinement for Unsupervised Cardiac MR Registration},
  author = {Yunguan Fu and Wenjia Bai and Wen Yan and Matthew J Clarkson and Rhodri Huw Davies and Yipeng Hu},
  year   = {2026},
  eprint = {2603.01073},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2603.01073v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2603.01073.svg)](https://codeofpaper.com/paper/2603.01073)

Discussion