提出区域拖拽编辑方法，实现精准形状变换。
In-context Region-based Drag: Drag Any Region to Any Shape
- 基于上下文学习，通过图像与掩码注意力一致性约束
- 在多个指标上优于现有方法，视觉保真度更高
- 适合需要精确区域编辑的研究者与开发者
扩散模型在拖拽式编辑中展现出潜力。以往工作主要关注点级拖拽，存在固有歧义。本文聚焦于区域级拖拽，提出一种新型的上下文区域拖拽（ICRDrag）方法。在上下文学习框架下，ICRDrag 输入源图像、源区域掩码和目标区域掩码，生成目标拖拽图像。在基础上下文学习模型基础上，引入两种新颖的注意力正则化：1）图像-掩码注意力一致性，确保目标区域在图像与掩码模态中关注相似的源区域；2）源-目标注意力对应性，保证源与目标区域间的相互对应。为支持区域级拖拽，还构建了大规模成对区域数据集（PRD），包含配对的掩码与图像。大量实验表明，ICRDrag 在定量指标和用户研究中均显著优于现有方法，编辑准确率与视觉保真度更优。代码、模型与数据集已开源。
原文摘要 · Abstract (English)
Diffusion models have shown promise in drag-style editing. Previous works mainly focus on point-based drag, which is inherently ambiguous. This paper focuses on region-based drag and introduces a novel In-Context Region-based Drag (ICRDrag) method. Under the in-context learning framework, ICRDrag consumes a source image, a source region mask, and a target region mask, producing the target dragged image. Built upon the basic in-context learning model, we introduce two novel attention regularization: 1) image-mask attention consistency to ensure that a target region attends to similar source regions for image and mask modalities; 2) source-target attention correspondence to ensure the mutual correspondence between source and target regions. To facilitate region-based drag, we also construct Paired Region Dataset (PRD), a large-scale dataset with paired masks and images. Extensive experiments show that ICRDrag significantly outperforms existing methods in both quantitative metrics and user studies, achieving superior editing accuracy and visual fidelity. The dataset, code, and model are available at https://github.com/bcmi/ICRDrag-Region-Drag-Editing.
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。