DreamMix：解耦对象属性以增强定制图像修复的可编辑性
DreamMix: Decoupling Object Attributes for Enhanced Editability in Customized Image Inpainting
AI总结：
DreamMix提出一种基于扩散模型的框架，通过属性解耦机制、文本属性替换和解耦修复框架，在定制图像修复中实现身份保持与属性可编辑性的平衡，支持对象插入、属性编辑和小对象修复。
摘要：
随着扩散模型的快速发展，主题驱动的图像修复在图像编辑领域近期受到广泛关注。除图像引导外，近期研究探索引入文本引导，以实现保持身份一致且局部可编辑的对象修复。然而，这些方法仍存在身份过拟合问题，即原始属性与目标文本指令纠缠不清。为克服此局限，我们提出DreamMix，一种基于扩散的框架，擅长将目标对象插入用户指定区域，同时支持任意文本驱动的属性修改。DreamMix引入三个关键组件：(i)属性解耦机制（ADM），合成多样化的属性增强图像-文本对以缓解过拟合；(ii)文本属性替换（TAS）模块，通过正交分解隔离目标属性；(iii)解耦修复框架（DIF），将局部生成与全局协调分离。在多种修复骨干网络上的大量实验表明，DreamMix在身份保持与属性可编辑性之间取得了优越平衡，适用于对象插入、属性编辑和小对象修复等多样化应用。
英文摘要：
Subject-driven image inpainting has recently gained prominence in image editing with the rapid advancement of diffusion models. Beyond image guidance, recent studies have explored incorporating text guidance to achieve identity-preserved yet locally editable object inpainting. However, these methods still suffer from identity overfitting, where original attributes remain entangled with target textual instructions. To overcome this limitation, we propose DreamMix, a diffusion-based framework adept at inserting target objects into user-specified regions while concurrently enabling arbitrary text-driven attribute modifications. DreamMix introduces three key components: (i) an Attribute Decoupling Mechanism (ADM) that synthesizes diverse attribute-augmented image-text pairs to mitigate overfitting; (ii) a Textual Attribute Substitution (TAS) module that isolates target attributes via orthogonal decomposition, and (iii) a Disentangled Inpainting Framework (DIF) that seperates local generation from global harmonization. Extensive experiments across multiple inpainting backbones demonstrate that DreamMix achieves a superior balance between identity preservation and attribute editability across diverse applications, including object insertion, attribute editing, and small object inpainting.
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。
