EDITBRIDGE：面向超高分辨率图像的保真高效编辑

摘要

高分辨率图像编辑在专业工作流中的需求日益增长，然而现有基于扩散的模型由于二次注意力复杂度和高昂的内存需求，仍受限于1K以下的分辨率。一种常见的替代方案采用两阶段流程：先进行低分辨率编辑，再进行独立的超分辨率处理。然而，该方法存在两个关键问题：信息偏离，即幻觉细节与原始高分辨率（HR）源图像相矛盾；以及纹理退化，表现为过度平滑或过度锐化的伪影。我们提出EditBridge，一种用于高效超高分辨率编辑的扩散桥框架。与从噪声重新生成的传统扩散方法不同，我们将精化过程建模为从低分辨率（LR）编辑结果到其高分辨率对应图像的结构化数据到数据转换，并显式以原始HR源图像为条件以保留真实细节。为高效融入HR源图像引导，我们引入了一种先验引导的分块稀疏注意力机制，利用第一阶段编辑中的语义对应关系将跨图像交互约束在空间对齐区域内，从而显著降低计算开销。大量实验表明，EditBridge在高达4K分辨率下实现了高保真编辑和卓越的感知质量，在2K分辨率下实现了3.6–8.4倍加速，并在61秒内实现了实用的4K编辑。

English

High-resolution image editing is increasingly demanded in professional workflows, yet existing diffusion-based models remain constrained to resolutions below 1K due to quadratic attention complexity and prohibitive memory requirements. A prevalent workaround employs a two-stage pipeline: editing at low resolution followed by independent super-resolution. However, this approach suffers from two critical issues: information divergence, where hallucinated details contradict the original high-resolution (HR) source, and texture degradation, manifesting as over-smoothed or over-sharpened artifacts. We propose EditBridge, a diffusion bridge framework for efficient ultra high-resolution editing. Unlike conventional diffusion that regenerates from noise, we formulate refinement as structured data-to-data translation from the low-resolution (LR) edited result to its HR counterpart, explicitly conditioned on the original HR source to preserve authentic details. To efficiently incorporate HR source guidance, we introduce a prior-guided block-wise sparse attention mechanism that exploits semantic correspondence from first-stage editing to constrain cross-image interactions to spatially aligned regions, significantly reducing computational overhead. Extensive experiments demonstrate that EditBridge achieves high-fidelity editing with superior perceptual quality at resolutions up to 4K, delivering 3.6--8.4times speedup at 2K and enabling practical 4K editing in 61 seconds.