{"ID":2879035,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2508.16852","arxiv_id":"2508.16852","title":"Gaussian Primitive Optimized Deformable Retinal Image Registration","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.","short_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 perfo...","url_abs":"https://arxiv.org/abs/2508.16852","url_pdf":"https://arxiv.org/pdf/2508.16852v1","authors":"[\"Xin Tian\",\"Jiazheng Wang\",\"Yuxi Zhang\",\"Xiang Chen\",\"Renjiu Hu\",\"Gaolei Li\",\"Min Liu\",\"Hang Zhang\"]","published":"2025-08-23T00:44:50Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\",\"eess.IV\"]","methods":"[]","has_code":false,"code_links":[{"ID":610543,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2879035,"paper_url":"https://arxiv.org/abs/2508.16852","paper_title":"Gaussian Primitive Optimized Deformable Retinal Image Registration","repo_url":"https://github.com/xintian-99/GPOreg","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}