提出新评估方法，精准衡量图像配准的对齐与平滑性权衡。
Evaluation of Deformable Image Registration under Alignment-Regularity Trade-off
- 用弧线连续刻画配准方法在不同平滑度下的表现
- 揭示传统评估忽略的性能隐藏规律
- 适合需要精细选择配准模型的研究者
可变形图像配准（DIR）的评估面临对齐精度与形变平滑性之间的固有权衡。现有方法大多未能妥善处理或完全忽视这一问题。本文指出当前实践缺陷，提出一种能持续捕捉该权衡的评估方案。首先引入对齐-正则性特征（ARC）曲线，描述给定配准方法在不同正则化程度下的性能谱。实验表明，ARC曲线揭示了传统评估无法体现的独特洞见，涵盖多种深度学习架构和变换模型的代表性方法。进一步采用基于HyperNetwork的方法，实现对全正则化范围的连续插值，加速了ARC曲线构建并提升了采样密度。最后，基于该评估框架，为从业者和研究人员提供细致的模型评价与选型指南。
原文摘要 · Abstract (English)
Evaluating deformable image registration (DIR) is challenging due to the inherent trade-off between achieving high alignment accuracy and maintaining deformation regularity. However, most existing DIR works either address this trade-off inadequately or overlook it altogether. In this paper, we highlight the issues with existing practices and propose an evaluation scheme that captures the trade-off continuously to holistically evaluate DIR methods. We first introduce the alignment regularity characteristic (ARC) curves, which describe the performance of a given registration method as a spectrum under various degrees of regularity. We demonstrate that the ARC curves reveal unique insights that are not evident from existing evaluation practices, using experiments on representative deep learning DIR methods with various network architectures and transformation models. We further adopt a HyperNetwork based approach that learns to continuously interpolate across the full regularization range, accelerating the construction and improving the sample density of ARC curves. Finally, we provide general guidelines for a nuanced model evaluation and selection using our evaluation scheme for both practitioners and registration researchers.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。