arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2301.04883cs.CLcs.CV
收藏

SlideVQA:一个多图像文档可视化问答数据集

SlideVQA: A Dataset for Document Visual Question Answering on Multiple Images

NTT Human Informatics Laboratories, NTT Corporation(日本电信电话株式会社人类信息学实验室)

Ryota Tanaka, Kyosuke Nishida, Kosuke Nishida, Taku Hasegawa, Itsumi Saito, Kuniko Saito

摘要:

包含文本、视觉和布局信息的文档图像上的视觉问答,称为文档VQA,近来受到了广泛的关注。尽管已经提出了许多数据集用于开发文档VQA系统,但大多数现有数据集都专注于理解单个图像内的内容关系,而不是多个图像之间的内容关系。在这项研究中,我们提出了一个新的多图像文档VQA数据集,SlideVQA,包含2.6k+幻灯片甲板组成的52k+幻灯片图像和14.5k问题的幻灯片甲板。SlideVQA需要复杂的推理,包括单跳、多跳和数值推理,并提供数值答案的注释算术表达式,以增强数值推理能力。此外,我们开发了一个新的端到端的文档VQA模型,将证据选择和问答在一个统一的序列到序列的格式。在SlideVQA上的实验表明,我们的模型优于现有的最先进的QA模型,但它仍然与人类的表现有很大的差距。我们相信,我们的数据集将促进对文件VQA的研究。

英文摘要:

Visual question answering on document images that contain textual, visual, and layout information, called document VQA, has received much attention recently. Although many datasets have been proposed for developing document VQA systems, most of the existing datasets focus on understanding the content relationships within a single image and not across multiple images. In this study, we propose a new multi-image document VQA dataset, SlideVQA, containing 2.6k+ slide decks composed of 52k+ slide images and 14.5k questions about a slide deck. SlideVQA requires complex reasoning, including single-hop, multi-hop, and numerical reasoning, and also provides annotated arithmetic expressions of numerical answers for enhancing the ability of numerical reasoning. Moreover, we developed a new end-to-end document VQA model that treats evidence selection and question answering in a unified sequence-to-sequence format. Experiments on SlideVQA show that our model outperformed existing state-of-the-art QA models, but that it still has a large gap behind human performance. We believe that our dataset will facilitate research on document VQA.

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。

↑