Cora：基於少步擴散的對應感知圖像編輯

摘要

影像編輯是電腦圖形學、視覺特效（VFX）領域中的一項重要任務，近期基於擴散模型的方法已能實現快速且高品質的成果。然而，對於需要大幅結構變更的編輯任務，如非剛性變形、物件修改或內容生成，仍面臨挑戰。現有的少步驟編輯方法常產生不相關的紋理等瑕疵，或難以保留源圖像的關鍵屬性（例如姿態）。我們提出了Cora，這是一種新穎的編輯框架，通過引入對應感知的噪聲校正和插值注意力圖來解決這些限制。我們的方法通過語義對齊源圖像與目標圖像間的紋理和結構，實現精確的紋理轉移，並在必要時生成新內容。Cora提供了在內容生成與保留之間平衡的控制。大量實驗表明，無論是定量還是定性分析，Cora在保持結構、紋理和身份識別方面均表現卓越，涵蓋姿態變化、物件添加及紋理精修等多種編輯場景。用戶研究證實，Cora提供的結果優於其他替代方案。

English

Image editing is an important task in computer graphics, vision, and VFX, with recent diffusion-based methods achieving fast and high-quality results. However, edits requiring significant structural changes, such as non-rigid deformations, object modifications, or content generation, remain challenging. Existing few step editing approaches produce artifacts such as irrelevant texture or struggle to preserve key attributes of the source image (e.g., pose). We introduce Cora, a novel editing framework that addresses these limitations by introducing correspondence-aware noise correction and interpolated attention maps. Our method aligns textures and structures between the source and target images through semantic correspondence, enabling accurate texture transfer while generating new content when necessary. Cora offers control over the balance between content generation and preservation. Extensive experiments demonstrate that, quantitatively and qualitatively, Cora excels in maintaining structure, textures, and identity across diverse edits, including pose changes, object addition, and texture refinements. User studies confirm that Cora delivers superior results, outperforming alternatives.

Cora：基於少步擴散的對應感知圖像編輯

Cora: Correspondence-aware image editing using few step diffusion

摘要

Support