EditID：面向文本到图像生成的免训练可编辑 ID 定制
EditID: Training-Free Editable ID Customization for Text-to-Image Generation
AI总结：
本文提出免训练的 EditID，基于 DiT 将定制 ID 生成解耦为图像与人物特征分支，通过映射、位移及整合强度控制兼顾 ID 一致性与可编辑性。
摘要：
我们提出 EditID，这是一种基于 DiT 架构的免训练方法，可在文本到图像生成中实现高度可编辑的定制化 ID。现有的定制 ID 文本到图像模型通常更关注 ID 一致性，却忽视可编辑性，因此难以通过提示改变面部朝向、人物属性及其他特征。EditID 通过将用于定制 ID 的文本到图像模型解构为图像生成分支和人物特征分支来解决该问题。人物特征分支进一步解耦为三个模块：特征提取、特征融合和特征整合。通过引入映射特征与位移特征的组合，并控制 ID 特征整合的强度，EditID 实现了跨网络深度的局部特征语义压缩，形成可编辑的特征空间。这使其能够成功生成具有可编辑 ID 的高质量图像，同时保持 ID 一致性，并在 IBench 评估中取得优异结果。IBench 是定制 ID 文本到图像生成领域的可编辑性评估框架，定量证明了 EditID 的优越性能。EditID 是首个在 DiT 架构上提出可定制 ID 可编辑性的文本到图像方案，可满足长提示和高质量图像生成的需求。
英文摘要：
We propose EditID, a training-free approach based on the DiT architecture, which achieves highly editable customized IDs for text to image generation. Existing text-to-image models for customized IDs typically focus more on ID consistency while neglecting editability. It is challenging to alter facial orientation, character attributes, and other features through prompts. EditID addresses this by deconstructing the text-to-image model for customized IDs into an image generation branch and a character feature branch. The character feature branch is further decoupled into three modules: feature extraction, feature fusion, and feature integration. By introducing a combination of mapping features and shift features, along with controlling the intensity of ID feature integration, EditID achieves semantic compression of local features across network depths, forming an editable feature space. This enables the successful generation of high-quality images with editable IDs while maintaining ID consistency, achieving excellent results in the IBench evaluation, which is an editability evaluation framework for the field of customized ID text-to-image generation that quantitatively demonstrates the superior performance of EditID. EditID is the first text-to-image solution to propose customizable ID editability on the DiT architecture, meeting the demands of long prompts and high quality image generation.
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。
