MagicBrush：一个用于指令引导图像编辑的人工标注数据集
MagicBrush: A Manually Annotated Dataset for Instruction-Guided Image Editing
AI总结：
针对现有文本引导图像编辑方法依赖噪声大的合成数据、效果不佳的问题，本文推出首个大规模人工标注的指令引导真实图像编辑数据集MagicBrush，覆盖多类编辑场景，实验证明其能提升模型效果，也指出现有基准与真实需求的差距。
摘要：
文本引导的图像编辑在日常生活中需求广泛，涵盖从个人使用到Photoshop等专业应用的各类场景。然而，现有方法要么是零样本方法，要么是在自动合成数据集上训练的，这类数据集包含大量噪声，因此在实际应用中仍需大量人工调试才能产出理想结果。为解决这一问题，我们推出MagicBrush（https://osu-nlp-group.github.io/MagicBrush/），这是首个用于指令引导真实图像编辑的大规模人工标注数据集，覆盖单轮、多轮、提供掩码、无掩码等多样化编辑场景。MagicBrush包含超过1万组人工标注的三元组（源图像、编辑指令、目标图像），可支持大规模文本引导图像编辑模型的训练。我们在MagicBrush上微调InstructPix2Pix模型，人类评估结果显示新模型生成的图像质量显著提升。我们进一步开展了大量实验，从定量、定性以及人类评估多个维度评估当前图像编辑基准模型。结果表明我们的数据集具有挑战性，也揭示出现有基准模型与真实编辑需求之间存在差距。
英文摘要：
Text-guided image editing is widely needed in daily life, ranging from personal use to professional applications such as Photoshop. However, existing methods are either zero-shot or trained on an automatically synthesized dataset, which contains a high volume of noise. Thus, they still require lots of manual tuning to produce desirable outcomes in practice. To address this issue, we introduce MagicBrush (https://osu-nlp-group.github.io/MagicBrush/), the first large-scale, manually annotated dataset for instruction-guided real image editing that covers diverse scenarios: single-turn, multi-turn, mask-provided, and mask-free editing. MagicBrush comprises over 10K manually annotated triplets (source image, instruction, target image), which supports trainining large-scale text-guided image editing models. We fine-tune InstructPix2Pix on MagicBrush and show that the new model can produce much better images according to human evaluation. We further conduct extensive experiments to evaluate current image editing baselines from multiple dimensions including quantitative, qualitative, and human evaluations. The results reveal the challenging nature of our dataset and the gap between current baselines and real-world editing needs.
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。
