无需训练的图像编辑方法，提升参考图像修改质量与效率
Reversible Inversion for Training-Free Exemplar-guided Image Editing
- 两阶段去噪反演，先源图后参考图，优化编辑过程
- 引入掩码引导选择性去噪，精准控制修改区域
- 相比现有方法性能更优，计算开销更低
示例引导图像编辑（EIE）旨在根据视觉参考修改源图像。现有方法通常需要大规模预训练以学习源图与参考图之间的关系，带来高计算成本。作为无需训练的替代方案，反演技术可将源图像映射到潜在空间进行操作。然而，我们的实证研究发现标准反演在EIE中表现不佳，导致质量差且效率低。为此，我们提出 extbf{可逆反演（ReInversion）}，实现高效、高质量的无训练EIE。具体而言，ReInversion采用两阶段去噪过程：第一阶段基于源图像，第二阶段基于参考图像。此外，引入掩码引导的选择性去噪（MSD）策略，限制修改区域，保持背景结构一致性。定性与定量对比表明，本方法在计算开销最低的前提下达到当前最优的EIE性能。
原文摘要 · Abstract (English)
Exemplar-guided Image Editing (EIE) aims to modify a source image according to a visual reference. Existing approaches often require large-scale pre-training to learn relationships between the source and reference images, incurring high computational costs. As a training-free alternative, inversion techniques can be used to map the source image into a latent space for manipulation. However, our empirical study reveals that standard inversion is sub-optimal for EIE, leading to poor quality and inefficiency. To tackle this challenge, we introduce \textbf{Reversible Inversion ({ReInversion})} for effective and efficient EIE. Specifically, ReInversion operates as a two-stage denoising process, which is first conditioned on the source image and subsequently on the reference. Besides, we introduce a Mask-Guided Selective Denoising (MSD) strategy to constrain edits to target regions, preserving the structural consistency of the background. Both qualitative and quantitative comparisons demonstrate that our ReInversion method achieves state-of-the-art EIE performance with the lowest computational overhead.
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。