轻量级网络实现跨域图像渐进式精准配准，提升医学影像对齐效果。
PCReg-Net: Progressive Contrast-Guided Registration for Cross-Domain Image Alignment
- 分阶段设计：从粗到精逐步优化，融合多尺度对比与特征注入。
- 仅256万参数，实时推理达141帧/秒，优于传统与深度学习方法。
- 适用于眼底、显微等跨域场景，代码开源可复现。
跨域可变形图像配准因外观变化与几何错位违反亮度恒常性假设而极具挑战。本文提出PCReg-Net，一种渐进式对比引导配准框架，包含四个轻量模块：(1) 注册U-Net实现初始粗略对齐；(2) 参考特征提取器从固定图像中捕捉多尺度结构信息；(3) 多尺度对比模块通过比较粗配准与参考特征识别残余错位；(4) 带特征注入的细化U-Net生成高保真输出。在FIRE-Reg-256眼底图像基准上验证，性能超越传统及深度学习基线。在两个显微镜基准上进一步验证其跨域适用性。模型仅含256万参数，实现141帧/秒实时推理。代码已开源：https://github.com/JiahaoQin/PCReg-Net。
原文摘要 · Abstract (English)
Deformable image registration across heterogeneous domains remains challenging because coupled appearance variation and geometric misalignment violate the brightness constancy assumption underlying conventional methods. We propose PCReg-Net, a progressive contrast-guided registration framework that performs coarse-to-fine alignment through four lightweight modules: (1)~a registration U-Net for initial coarse alignment, (2)~a reference feature extractor capturing multi-scale structural cues from the fixed image, (3)~a multi-scale contrast module that identifies residual misalignment by comparing coarse-registered and reference features, and (4)~a refinement U-Net with feature injection that produces the final high-fidelity output. We evaluate on the FIRE-Reg-256 retinal fundus benchmark, demonstrating improvements over both traditional and deep learning baselines. Additional experiments on two microscopy benchmarks further confirm cross-domain applicability. With only 2.56M parameters, PCReg-Net achieves real-time inference at 141 FPS. Code is available at https://github.com/JiahaoQin/PCReg-Net.
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。