以最少文本输入实现用户友好的图像编辑：利用字幕生成与注入技术
User-friendly Image Editing with Minimal Text Input: Leveraging Captioning and Injection Techniques
AI总结：
针对文本驱动图像编辑中提示词工程繁琐的问题，该研究结合提示词生成、字幕生成与注入技术，使用户仅需极少文本即可获得接近真值提示词的编辑效果。
摘要：
近年来，扩散模型中的文本驱动图像编辑已取得显著成功。然而，现有方法假设用户的描述能够充分锚定源图像中的上下文，例如物体、背景、风格及其相互关系。这一假设并不适用于真实世界应用，因为用户必须手动设计文本提示词，才能为不同图像找到最优描述。从用户角度来看，提示词工程是一项劳动密集型过程，用户更倾向于提供一个用于编辑的目标词，而不是完整句子。为解决该问题，我们首先根据语义细节的层级将提示词划分为三类，以此证明对源图像进行详细文本描述的重要性。随后，我们通过组合提示词生成框架，提出了简单而有效的方法，从而使提示词工程过程更加用户友好。大量定性与定量实验证明了提示词在文本驱动图像编辑中的重要性，并且我们的方法可与真值提示词相媲美。
英文摘要：
Recent text-driven image editing in diffusion models has shown remarkable success. However, the existing methods assume that the user's description sufficiently grounds the contexts in the source image, such as objects, background, style, and their relations. This assumption is unsuitable for real-world applications because users have to manually engineer text prompts to find optimal descriptions for different images. From the users' standpoint, prompt engineering is a labor-intensive process, and users prefer to provide a target word for editing instead of a full sentence. To address this problem, we first demonstrate the importance of a detailed text description of the source image, by dividing prompts into three categories based on the level of semantic details. Then, we propose simple yet effective methods by combining prompt generation frameworks, thereby making the prompt engineering process more user-friendly. Extensive qualitative and quantitative experiments demonstrate the importance of prompts in text-driven image editing and our method is comparable to ground-truth prompts.
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。
