无需微调，可精准迁移复杂纹理如云、火的图像编辑方法
TextureDiffusion: Target Prompt Disentangled Editing for Various Texture Transfer
- 将目标提示设为纯纹理词，实现内容与纹理解耦
- 利用自注意力和残差块特征保留原图结构，背景不变
- 适合需要高质量纹理迁移的创意设计与图像生成场景
近期文本引导图像编辑取得显著进展，但现有方法仅能处理木纹、金箔等简单纹理，复杂纹理如云或火焰仍具挑战。根源在于目标提示需同时包含输入图像内容与<纹理>，限制了纹理表征能力。本文提出TextureDiffusion，一种无需微调的图像编辑方法，适用于多种纹理迁移。首先，将目标提示直接设为<texture>，使纹理与输入内容解耦，增强纹理表达；其次，利用自注意力中的查询特征和残差块特征，保留输入图像结构；最后，引入编辑定位技术，融合自注意力结果与中间潜在表示，维持背景完整性。大量实验表明，TextureDiffusion能和谐迁移各类纹理，结构与背景保持优异。代码已公开于https://github.com/THU-CVML/TextureDiffusion。
原文摘要 · Abstract (English)
Recently, text-guided image editing has achieved significant success. However, existing methods can only apply simple textures like wood or gold when changing the texture of an object. Complex textures such as cloud or fire pose a challenge. This limitation stems from that the target prompt needs to contain both the input image content and <texture>, restricting the texture representation. In this paper, we propose TextureDiffusion, a tuning-free image editing method applied to various texture transfer. Initially, the target prompt is directly set to "<texture>", making the texture disentangled from the input image content to enhance texture representation. Subsequently, query features in self-attention and features in residual blocks are utilized to preserve the structure of the input image. Finally, to maintain the background, we introduce an edit localization technique which blends the self-attention results and the intermediate latents. Comprehensive experiments demonstrate that TextureDiffusion can harmoniously transfer various textures with excellent structure and background preservation. Code is publicly available at https://github.com/THU-CVML/TextureDiffusion
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。