Gaussian Primitive Optimized Deformable Retinal Image Registration

TL;DR

This paper introduces Gaussian Primitive Optimization (GPO), a novel iterative framework for deformable retinal image registration that leverages keypoints and Gaussian primitives to improve accuracy and robustness.

Contribution

GPO is a new structured message passing framework that models keypoints as Gaussian primitives and optimizes registration end-to-end, addressing challenges in retinal image registration.

Findings

Reduces registration error from 6.2 px to 2.4 px

Increases AUC at 25 px from 0.770 to 0.938

Outperforms existing methods on FIRE dataset

Abstract

Deformable retinal image registration is notoriously difficult due to large homogeneous regions and sparse but critical vascular features, which cause limited gradient signals in standard learning-based frameworks. In this paper, we introduce Gaussian Primitive Optimization (GPO), a novel iterative framework that performs structured message passing to overcome these challenges. After an initial coarse alignment, we extract keypoints at salient anatomical structures (e.g., major vessels) to serve as a minimal set of descriptor-based control nodes (DCN). Each node is modelled as a Gaussian primitive with trainable position, displacement, and radius, thus adapting its spatial influence to local deformation scales. A K-Nearest Neighbors (KNN) Gaussian interpolation then blends and propagates displacement signals from these information-rich nodes to construct a globally coherent displacement field; focusing interpolation on the top (K) neighbors reduces computational overhead while preserving local detail. By strategically anchoring nodes in high-gradient regions, GPO ensures robust gradient flow, mitigating vanishing gradient signal in textureless areas. The framework is optimized end-to-end via a multi-term loss that enforces both keypoint consistency and intensity alignment. Experiments on the FIRE dataset show that GPO reduces the target registration error from 6.2\,px to ~2.4\,px and increases the AUC at 25\,px from 0.770 to 0.938, substantially outperforming existing methods. The source code can be accessed via https://github.com/xintian-99/GPOreg.