PCReg-Net: Progressive Contrast-Guided Registration for Cross-Domain Image Alignment

TL;DR

PCReg-Net is a lightweight, progressive registration framework that effectively aligns cross-domain images by combining coarse-to-fine strategies and multi-scale contrast features, achieving real-time performance.

Contribution

The paper introduces PCReg-Net, a novel progressive contrast-guided registration method with multi-scale feature extraction and refinement modules for cross-domain image alignment.

Findings

Outperforms traditional and deep learning baselines on retinal and microscopy benchmarks.

Achieves real-time inference at 141 FPS with only 2.56 million parameters.

Demonstrates cross-domain applicability of the registration framework.

Abstract

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.