arXivDaily arXiv每日学术速递 周一至周五更新
​通知:联邦节假日调休:哥伦布日 / 原住民日(Columbus Day / Indigenous Peoples' Day),今日(13日)无更新。
arXiv 2503.04268cs.CV

ControlFill:基于提示学习的空间可调图像修复

ControlFill: Spatially Adjustable Image Inpainting from Prompt Learning

Samsung Research(三星研究院)

Boseong Jeon

AI总结:

提出ControlFill图像修复框架,通过训练creation和removal两种提示引导无需重型文本编码器的扩散网络,并结合无分类器引导与像素级尺度分配实现空间可调的修复强度控制。

摘要:

在本报告中,我提出了一个名为ControlFill的图像修复框架,该框架涉及训练两种不同的提示:一种用于在指定掩码内生成合理的对象(creation),另一种用于通过扩展背景来填充该区域(removal)。在推理阶段,这些学习到的嵌入向量会引导一个无需重型文本编码器的扩散网络。通过调整这两个提示的相对重要性并采用无分类器引导,用户可以控制removal或creation的强度。此外,我还引入了一种方法,通过为单个像素分配不同的尺度,在空间上改变引导的强度。

英文摘要:

In this report, I present an inpainting framework named \textit{ControlFill}, which involves training two distinct prompts: one for generating plausible objects within a designated mask (\textit{creation}) and another for filling the region by extending the background (\textit{removal}). During the inference stage, these learned embeddings guide a diffusion network that operates without requiring heavy text encoders. By adjusting the relative significance of the two prompts and employing classifier-free guidance, users can control the intensity of removal or creation. Furthermore, I introduce a method to spatially vary the intensity of guidance by assigning different scales to individual pixels.

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。

↑