注意力范围引导:用于图像编辑的无训练空间控制
Attention-Scoped Guidance: Training-Free Spatial Control for Image Editing
- Shanghai Jiao Tong University(上海交通大学)
- Cardiff University(卡迪夫大学)
- University of Klagenfurt(克拉根福大学)
机构由 AI 辅助整理,请以论文原文为准。
AI总结:
提出注意力范围引导(ASG),一种无训练的空间权重分配方法,通过指令注意力软支持图调节文本引导与图像锚定,在MagicBrush和PIE-Bench++上提升编辑保留指标。
AI中文摘要:
指令引导的图像编辑应改变指令所指定的内容,并保持图像其余部分不变。在双分类器自由引导(CFG)中,编辑器在每次去噪步骤中结合两个方向:一个方向推动向指令编辑,另一个方向拉回源图像,并使用全局权重。我们引入了注意力范围引导(ASG),一种采样器包装器,使这些权重具有空间性。它从编辑器已计算的指令注意力中读取软支持图,然后在支持度低的地方减弱文本引导,在支持度高的地方加强图像锚定。该包装器无需训练、无需外部掩码、无需额外的网络评估。在完整的MagicBrush和PIE-Bench++分割上,ASG改善了保留导向的指标,在MagicBrush的四个指标中领先三个,并改善了PIE-Bench++的背景PSNR。一个剂量匹配的对照(去除空间放置)在PIE-Bench++上损失高达0.73 CLIP,证实了空间分配本身带来了增益。
英文摘要:
Instruction-guided image editing should change what the instruction names and leave the rest of the image untouched. In dual classifier-free guidance (CFG), an editor combines two directions at every denoising step, one that pushes toward the instructed edit and one that pulls back toward the source image, using global weights. We introduce Attention-Scoped Guidance (ASG), a sampler wrapper that makes these weights spatial. It reads a soft support map from the instruction attention that the editor already computes, then weakens text guidance where support is low and strengthens image anchoring where support is high. The wrapper requires no training, no external mask, and no additional network evaluation. On the full MagicBrush and PIE-Bench++ splits, ASG improves preservation-oriented metrics, leading three of four MagicBrush metrics and PIE-Bench++ background PSNR. A dose-matched control that removes the spatial placement loses up to 0.73 CLIP on PIE-Bench++, confirming that the spatial allocation itself carries the gain.