RelationAdapter：利用扩散Transformer学习和迁移视觉关系
RelationAdapter: Learning and Transferring Visual Relation with Diffusion Transformers
AI总结：
本文提出轻量级模块RelationAdapter和数据集Relation252K，使扩散Transformer能从源-目标图像对中学习并迁移视觉编辑关系，显著提升视觉提示驱动图像编辑的泛化能力与生成质量。
摘要：
受大语言模型（LLMs）的上下文学习机制启发，一种基于可泛化视觉提示的图像编辑新范式正在兴起。现有的单参考方法通常侧重于风格或外观调整，难以处理非刚性变换。为解决这些局限，我们提出利用源-目标图像对来提取内容感知的编辑意图，并将其迁移到新的查询图像。为此，我们引入了RelationAdapter，这是一个轻量级模块，使基于扩散Transformer（DiT）的模型能够从极少量示例中有效捕获并应用视觉变换。我们还引入了Relation252K，一个包含218种多样化编辑任务的综合数据集，用于评估模型在视觉提示驱动场景下的泛化能力和适应性。在Relation252K上的实验表明，RelationAdapter显著提升了模型理解和迁移编辑意图的能力，从而在生成质量和整体编辑性能上取得了显著提升。
英文摘要：
Inspired by the in-context learning mechanism of large language models (LLMs), a new paradigm of generalizable visual prompt-based image editing is emerging. Existing single-reference methods typically focus on style or appearance adjustments and struggle with non-rigid transformations. To address these limitations, we propose leveraging source-target image pairs to extract and transfer content-aware editing intent to novel query images. To this end, we introduce RelationAdapter, a lightweight module that enables Diffusion Transformer (DiT) based models to effectively capture and apply visual transformations from minimal examples. We also introduce Relation252K, a comprehensive dataset comprising 218 diverse editing tasks, to evaluate model generalization and adaptability in visual prompt-driven scenarios. Experiments on Relation252K show that RelationAdapter significantly improves the model's ability to understand and transfer editing intent, leading to notable gains in generation quality and overall editing performance.
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。
