让图像修复时物体属性可自由编辑，且不丢失原物身份
DreamMix: Decoupling Object Attributes for Enhanced Editability in Customized Image Inpainting
- 通过解耦属性与对象，实现文本控制下的任意属性修改
- 在多种任务中同时保持身份一致性和局部编辑能力
- 适合需要精准控制图像元素属性的创意设计场景
基于扩散模型的主体驱动图像修复近年来发展迅速。现有方法虽引入文本指导实现保留身份的局部编辑，但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。为此，我们提出DreamMix，一种基于扩散模型的框架，可在指定区域插入目标物体，并支持任意文本驱动的属性修改。其核心包含三个组件：(i) 属性解耦机制（ADM），生成多样化的属性增强图文对以缓解过拟合；(ii) 文本属性替换模块（TAS），通过正交分解分离目标属性；(iii) 解耦修复框架（DIF），将局部生成与全局调和过程分离。在多个修复主干网络上的实验表明，DreamMix在对象插入、属性编辑及小物体修复等任务中，均实现了身份保留与属性可编辑性之间的更优平衡。
原文摘要 · Abstract (English)
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.
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。