基于集合的文本到图像生成的离线评估
Offline Evaluation of Set-Based Text-to-Image Generation
AI总结：
针对文本到图像系统在创意构思场景中的评估问题，本文提出了一组基于用户浏览与交互模型的离线评估指标，综合考虑生成图像的相关性、多样性和排列方式，并通过人类研究验证了以用户使用方式为基础设计基准的重要性。
摘要：
文本到图像（TTI）系统通常在构思阶段为人们提供支持，构思是创意过程的早期阶段，此时接触广泛的相关图像集合有助于探索设计空间。由于构思是TTI任务的一个重要子类，理解如何根据TTI系统对构思的支持程度来定量评估它们，对于促进面向这些用户的研究与开发至关重要。然而，现有的TTI评估指标仍然侧重于分布相似性指标，如Fréchet Inception Distance（FID）。我们采取了一种替代方法，基于排序评估中的既有方法，开发了具有显式用户浏览和交互模型的TTI评估指标，这些模型刻画了用户如何浏览空间排列的生成图像集合并与之交互。我们提出的TTI离线评估指标不仅捕捉生成图像与用户构思需求的相关程度，还考虑了生成图像集合的多样性和排列方式。我们通过在图像网格上的人类研究来分析所提出的TTI指标族，这些图像网格由三个不同的TTI系统基于广泛使用的基准（如MS-COCO captions和Localized Narratives）的子集以及自然场景中使用的提示生成。我们的结果表明，将指标建立在人们如何使用系统的基础上，是基准设计中一个重要但尚未得到充分研究的领域。
英文摘要：
Text-to-Image (TTI) systems often support people during ideation, the early stages of a creative process when exposure to a broad set of relevant images can help explore the design space. Since ideation is an important subclass of TTI tasks, understanding how to quantitatively evaluate TTI systems according to how well they support ideation is crucial to promoting research and development for these users. However, existing evaluation metrics for TTI remain focused on distributional similarity metrics like Fréchet Inception Distance (FID). We take an alternative approach and, based on established methods from ranking evaluation, develop TTI evaluation metrics with explicit models of how users browse and interact with sets of spatially arranged generated images. Our proposed offline evaluation metrics for TTI not only capture how relevant generated images are with respect to the user's ideation need but also take into consideration the diversity and arrangement of the set of generated images. We analyze our proposed family of TTI metrics using human studies on image grids generated by three different TTI systems based on subsets of the widely used benchmarks such as MS-COCO captions and Localized Narratives as well as prompts used in naturalistic settings. Our results demonstrate that grounding metrics in how people use systems is an important and understudied area of benchmark design.
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。
