arXiv cs.AI· Chenyu Guo, Wenjie Dai, Yiqian Chang, Zhiyuan Zhang, Yunfan Wei, Shuhan Qi·· 2 小时前AI 评分26
ReliMorph：基于可靠性引导的专家路由与速度门控 3D 形变配准框架
Learning Where to Trust: Reliability-Guided Expert Routing and Velocity Gating for 3D Deformable Registration
AI 导读
研究者提出 ReliMorph，一个可靠性路由的对称 3D 形变图像配准框架，将对应关系可靠性评估与空间专家融合及残差速度门控耦合。该方法利用相关性分布刻画匹配模糊性，并在粗到细精化中通过逆不一致反馈引导路由与门控。据称这是首个在 3D 形变配准中同时以可靠性条件化空间专家路由和残差速度门控的框架，实验显示其在单模态与多模态配准任务的解剖重叠度上优于竞争方法。
正文 · 原文
Abstract:Deformable image registration relies on accurate anatomical correspondences. However, explicitly coupling local matching reliability with spatial context selection and deformation update strength remains insufficiently explored in learning-based registration. This challenge is particularly pronounced in homogeneous, weakly textured, and severely deformed regions, where ambiguous matches may induce erroneous deformation updates. We propose ReliMorph, a reliability-routed symmetric framework for 3D deformable registration that couples correspondence assessment with spatial context selection and update this http URL correlation distributions characterize correspondence ambiguity using shared anatomical features with modality-specific adaptation. These reliability cues guide spatial expert fusion and gate residual velocities during coarse-to-fine refinement. Updated inverse-inconsistency feedback guides routing and gating at finer pyramid levels. To our knowledge, ReliMorph is the first 3D deformable registration framework to jointly condition spatial expert routing and residual velocity gating on correspondence reliability. Extensive experiments demonstrate that ReliMorph consistently outperforms competing methods in anatomical overlap, highlighting its effectiveness across unimodal and multimodal registration tasks.
| Subjects: | Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.36672 [cs.AI] |
| (or arXiv:2610.36672v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.36672 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Chenyu Guo [view email]
[v1]
Tue, 27 Oct 2026 05:43:45 UTC (9,286 KB)
来源：arXiv cs.AI · arxiv.org