SAS-Net：通过结构-外观分解实现跨域图像配准作为逆渲染
SAS-Net: Cross-Domain Image Registration as Inverse Rendering via Structure-Appearance Factorization
AI总结：
本文提出SAS-Net，通过结构-外观分解解决跨域图像配准问题，采用逆渲染方法实现跨域对齐，实验表明其在卫星遥感和视网膜成像领域表现优异。
摘要：
跨域图像配准需要对异构成像物理获取的图像进行对齐，其中经典亮度恒定假设被根本性破坏。我们通过图像形成模型I = R(s, a) + ε来建模，其中每个观测由渲染函数R作用于领域不变场景结构s和领域特定外观统计a生成。配准则转化为逆渲染问题：给定两个领域的观测，恢复共享结构并重新渲染以获得配准输出。我们将其框架命名为SAS-Net（场景-外观分离网络），其中实例归一化实现结构-外观分解，自适应实例归一化（AdaIN）实现可微向前渲染器。场景一致性损失在分解的潜在空间中强制几何对应。在EuroSAT-Reg-256（卫星遥感）和FIRE-Reg-256（视网膜底片）上的实验表明，SAS-Net在异构成像领域实现了最先进的性能。SAS-Net（3.35M参数）在RTX 5090 GPU上达到89 FPS。代码：https://github.com/D-ST-Sword/SAS-Net。
英文摘要：
Cross-domain image registration requires aligning images acquired under heterogeneous imaging physics, where the classical brightness constancy assumption is fundamentally violated. We formulate this problem through an image formation model I = R(s, a) + epsilon, where each observation is generated by a rendering function R acting on domain-invariant scene structure s and domain-specific appearance statistics a. Registration then reduces to an inverse rendering problem: given observations from two domains, recover the shared structure and re-render it under the target appearance to obtain the registered output. We instantiate this framework as SAS-Net (Scene-Appearance Separation Network), where instance normalization implements the structure-appearance decomposition and Adaptive Instance Normalization (AdaIN) realizes the differentiable forward renderer. A scene consistency loss enforces geometric correspondence in the factorized latent space. Experiments on EuroSAT-Reg-256 (satellite remote sensing) and FIRE-Reg-256 (retinal fundus) demonstrate state-of-the-art performance across heterogeneous imaging domains. SAS-Net (3.35M parameters) achieves 89 FPS on an RTX 5090 GPU. Code: https://github.com/D-ST-Sword/SAS-Net.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
