HQ-Edit：一个面向基于指令的图像编辑的高质量数据集
HQ-Edit: A High-Quality Dataset for Instruction-based Image Editing
AI总结：
本研究推出含约20万样本的高质量指令式图像编辑数据集HQ-Edit，提出基于GPT-4V和DALL-E 3的可扩展构建流水线与两项量化评估指标，经其微调的InstructPix2Pix性能达SOTA。
摘要：
本研究推出HQ-Edit，这是一个包含约20万条编辑样本的高质量基于指令的图像编辑数据集。与以往依赖属性引导或人工反馈构建数据集的方法不同，我们设计了一套可扩展的数据收集流水线，借助先进的基础模型即GPT-4V和DALL-E 3实现。为确保数据集的高质量，研究首先在线收集多样化样本并进行扩充，再用这些样本生成包含输入输出图像与详细文本提示的高质量双联图，随后通过后处理保障二者的精准对齐。此外，我们提出了对齐度（Alignment）和连贯性（Coherence）两项评估指标，利用GPT-4V对图像编辑对的质量进行量化评估。HQ-Edit的高分辨率图像细节丰富，且配有全面的编辑提示，大幅提升了现有图像编辑模型的性能。例如，经HQ-Edit微调的InstructPix2Pix可取得最先进的图像编辑效果，甚至超越了用人工标注数据微调的模型。项目页面为https://thefllood.github.io/HQEdit_web。
英文摘要：
This study introduces HQ-Edit, a high-quality instruction-based image editing dataset with around 200,000 edits. Unlike prior approaches relying on attribute guidance or human feedback on building datasets, we devise a scalable data collection pipeline leveraging advanced foundation models, namely GPT-4V and DALL-E 3. To ensure its high quality, diverse examples are first collected online, expanded, and then used to create high-quality diptychs featuring input and output images with detailed text prompts, followed by precise alignment ensured through post-processing. In addition, we propose two evaluation metrics, Alignment and Coherence, to quantitatively assess the quality of image edit pairs using GPT-4V. HQ-Edits high-resolution images, rich in detail and accompanied by comprehensive editing prompts, substantially enhance the capabilities of existing image editing models. For example, an HQ-Edit finetuned InstructPix2Pix can attain state-of-the-art image editing performance, even surpassing those models fine-tuned with human-annotated data. The project page is https://thefllood.github.io/HQEdit_web.
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。
