register images

This work addresses the limited physical interpretability and fine-grained deformation control in existing deep learning–based medical image registration methods. Inspired by the Demons algorithm, it introduces local discrepancy modeling as a physically grounded prior into the registration framework for the first time. A neural Demons layer generates anatomically consistent driving forces in latent feature space, while an attention mechanism coupled with variational registration enables dynamic force–displacement interaction. By decoupling discrepancy modeling from the deformation process, the approach enhances modularity and interpretability. Evaluated on multiple 3D brain MRI datasets, the method outperforms state-of-the-art learning-based and optimization-based approaches, with visualizations and statistical analyses confirming a high degree of alignment between the computed driving forces and actual deformations.

This study addresses the challenges of correspondence ambiguity caused by local geometric similarity and the erroneous exclusion of correct solutions during region matching in non-rigid point cloud registration. To overcome these issues, we propose CoCo-Reg, which decouples regional context from dense matching. Specifically, the method enriches point features via regional patches while preserving the global search space to avoid hard constraints, and optimizes patch similarity through identity-corrected overlap supervision. Experimental results demonstrate that CoCo-Reg reduces the mean correspondence error to 0.0547, achieving a 72.6% improvement over the baseline. Furthermore, the proportion of high-error points decreases significantly from 47.3% to 17.3%, indicating substantial improvements in both registration accuracy and robustness.

Medical image registration faces challenges in modeling large deformations and capturing long-range voxel correspondences. To address these issues, this paper proposes a recursive dynamic correlation registration framework. Methodologically, it introduces local correlation feature computation coupled with a recurrent dynamic search mechanism, incorporates a memory-augmented lightweight recurrent update module, and decouples motion and texture features to achieve efficient, low-redundancy voxel-level matching. Its key innovation lies in recursively adjusting the matching receptive field to jointly optimize global deformation modeling and local alignment accuracy. Evaluated on OASIS (non-affine), brain MRI, and abdominal CT datasets, the method achieves state-of-the-art performance: it reduces FLOPs by 90.5% and inference time by 96% compared to RDP, while simultaneously improving registration accuracy.

To address the poor robustness, low data efficiency, and weak interpretability in deformable image registration, this paper proposes a mesh-free registration framework based on geometric deep learning. The method formulates deformation modeling within a Lagrangian reference frame—enabling multi-resolution iterative optimization without intermediate resampling for the first time. A graph neural network dynamically updates node coordinates, adaptively reconstructs neighborhoods, and incorporates geometric priors to directly learn sparse, high-dimensional deformation fields in Euclidean space. This design significantly reduces modeling errors under large deformations. Evaluated on cross-subject brain MRI and respiratory-phase lung CT registration tasks, the approach achieves state-of-the-art performance, demonstrating superior accuracy, strong generalization across domains, and physically grounded interpretability.

This work addresses the challenge of high-fidelity, temporally consistent 3D motion reconstruction of non-rigid objects—such as clothed humans—from sparse, unstructured, and partially occluded observations. We propose the first end-to-end jointly optimized framework integrating implicit neural fields with explicit mesh deformation. Our method introduces face-level, differential-geometry-driven near-isometric constraints to enforce spatiotemporal consistency in deformations. It further incorporates temporal feature fusion and differentiable rendering optimized by monocular depth video, eliminating reliance on predefined parametric human models. Evaluated on human and animal motion reconstruction tasks, our approach significantly outperforms state-of-the-art methods, achieving substantial improvements in geometric detail preservation and temporal smoothness. Notably, it is the first method to enable high-quality non-rigid motion reconstruction solely from monocular depth video.

This study addresses the performance degradation in continuous image registration caused by mismatches between implicit deformation priors and target motion patterns. We systematically investigate the impact of parameterization methods on registration accuracy by comparing SIREN, B-splines, and multi-resolution strategies. Consequently, we propose an "implicit prior-deformation matching" design principle that elucidates the applicability of different parameterizations. Building upon this principle, the proposed MR-D-BSCP method achieves state-of-the-art performance in both brain MRI and lung CT registration tasks. These results effectively validate the critical role of aligning implicit priors with deformation characteristics to enhance medical image registration accuracy, providing practical guidance for the design of continuous registration models.

Non-rigid registration of soft-tissue point clouds under large deformations, noise, and outliers remains challenging due to unreliable correspondences and globally implausible deformations when relying solely on local distance metrics. This work proposes DINE, the first framework to incorporate learned global deformation statistical priors into point cloud registration. DINE employs a two-stage optimization strategy: it first pretrains a base model using Chamfer distance and then jointly optimizes the deformation field via maximum a posteriori estimation, integrating either PCA-based Gaussian or normalizing flow priors within the Robust-DefReg and DefTransNet backbones. Experiments demonstrate that DINE reduces Chamfer distance by 27–69% on the DeformedTissue dataset, improves robustness to outliers and Gaussian noise by 66% and 83%, respectively, and achieves performance gains of 59–79% under extreme deformation scenarios in SynBench.

Existing methods for non-rigid mesh sequence registration often suffer from reliance on per-instance optimization, limited category generalization, support for only pairwise inputs, or incomplete output correspondences. This work proposes a feed-forward neural network that leverages a topology-aware point representation and a multimodal encoding scheme to construct a global motion embedding, enabling a lightweight decoder to directly predict vertex deformations. The approach achieves efficient and complete multi-frame registration without iterative optimization and inherently supports motion interpolation at arbitrary time steps as well as high-quality mesh morphing. It demonstrates state-of-the-art performance across diverse object categories and complex non-rigid motions.

This work addresses the challenges in multimodal image registration, where modality-specific information often leaks into the shared feature space and existing methods struggle to jointly model global rigid alignment and local non-rigid deformations. To overcome these limitations, the authors propose HRNet, which employs a shared backbone enhanced with modality-specific batch normalization (MSBN) and introduces a cross-scale decoupling and adaptive projection module (CDAP) to effectively suppress modality interference. Furthermore, a hybrid parameter prediction module (HPPM) is designed to unify the prediction of rigid transformations and non-rigid deformation fields within an end-to-end, non-iterative framework. The proposed method achieves state-of-the-art performance in both rigid and non-rigid registration across four multimodal datasets.

提出了一种基于互信息标准的概率医学图像配准模型BINDER，通过假设潜在体素对应关系，使用优化和MCMC采样技术解决非线性和多模态图像配准问题。