解决扩散模型图像编辑中的文本属性泄漏问题，让修改更精准。
Addressing Text Embedding Leakage in Diffusion-based Image Editing
- 用对象限定嵌入分离文本特征，避免无关区域被影响
- 在多种编辑场景下，属性泄漏率显著降低
- 适合需要精确控制图像编辑的AI设计师和研究者
基于生成式扩散模型的文本图像编辑通过自然语言提示简化了传统工作流程。然而，当前方法仍存在关键问题：属性泄漏，即针对特定对象的编辑意外影响无关区域或其他目标对象。分析表明，根源在于自回归文本编码器生成的序列结尾（EOS）嵌入中固有的语义纠缠，会无差别聚合提示中的属性。为此，我们提出无属性泄漏编辑框架（ALE），从源头解决该问题。ALE结合对象受限嵌入（ORE）以解耦文本嵌入、区域引导混合交叉注意力掩码（RGB-CAM）实现空间精确注意力，以及背景融合（BB）以保留非编辑内容。为量化评估不同编辑方法的属性泄漏，我们构建了属性泄漏评估基准（ALE-Bench），包含全面的编辑场景与新指标。大量实验表明，ALE大幅降低属性泄漏，实现准确、多对象、文本驱动的图像编辑，同时忠实保留非目标内容。
原文摘要 · Abstract (English)
Text-based image editing, powered by generative diffusion models, lets users modify images through natural-language prompts and has dramatically simplified traditional workflows. Despite these advances, current methods still suffer from a critical problem: attribute leakage, where edits meant for specific objects unintentionally affect unrelated regions or other target objects. Our analysis reveals the root cause as the semantic entanglement inherent in End-of-Sequence (EOS) embeddings generated by autoregressive text encoders, which indiscriminately aggregate attributes across prompts. To address this issue, we introduce Attribute-Leakage-free Editing (ALE), a framework that tackles attribute leakage at its source. ALE combines Object-Restricted Embeddings (ORE) to disentangle text embeddings, Region-Guided Blending for Cross-Attention Masking (RGB-CAM) for spatially precise attention, and Background Blending (BB) to preserve non-edited content. To quantitatively evaluate attribute leakage across various editing methods, we propose the Attribute-Leakage Evaluation Benchmark (ALE-Bench), featuring comprehensive editing scenarios and new metrics. Extensive experiments show that ALE reduces attribute leakage by large margins, thereby enabling accurate, multi-object, text-driven image editing while faithfully preserving non-target content.
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。