Variational Geometry-aware Neural Network based Method for Solving High-dimensional Diffeomorphic Mapping Problems

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Zhiwen, Ho, Cheuk Hin, Lui, Lok Ming
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908626015223808
author Li, Zhiwen
Ho, Cheuk Hin
Lui, Lok Ming
author_facet Li, Zhiwen
Ho, Cheuk Hin
Lui, Lok Ming
contents Traditional methods for high-dimensional diffeomorphic mapping often struggle with the curse of dimensionality. We propose a mesh-free learning framework designed for $n$-dimensional mapping problems, seamlessly combining variational principles with quasi-conformal theory. Our approach ensures accurate, bijective mappings by regulating conformality distortion and volume distortion, enabling robust control over deformation quality. The framework is inherently compatible with gradient-based optimization and neural network architectures, making it highly flexible and scalable to higher-dimensional settings. Numerical experiments on both synthetic and real-world medical image data validate the accuracy, robustness, and effectiveness of the proposed method in complex registration scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2511_01911
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Variational Geometry-aware Neural Network based Method for Solving High-dimensional Diffeomorphic Mapping Problems
Li, Zhiwen
Ho, Cheuk Hin
Lui, Lok Ming
Machine Learning
Artificial Intelligence
Numerical Analysis
Differential Geometry
Traditional methods for high-dimensional diffeomorphic mapping often struggle with the curse of dimensionality. We propose a mesh-free learning framework designed for $n$-dimensional mapping problems, seamlessly combining variational principles with quasi-conformal theory. Our approach ensures accurate, bijective mappings by regulating conformality distortion and volume distortion, enabling robust control over deformation quality. The framework is inherently compatible with gradient-based optimization and neural network architectures, making it highly flexible and scalable to higher-dimensional settings. Numerical experiments on both synthetic and real-world medical image data validate the accuracy, robustness, and effectiveness of the proposed method in complex registration scenarios.
title Variational Geometry-aware Neural Network based Method for Solving High-dimensional Diffeomorphic Mapping Problems
topic Machine Learning
Artificial Intelligence
Numerical Analysis
Differential Geometry
url https://arxiv.org/abs/2511.01911