| Presentation | 2023-09-21 BoxPlotQA: Visual Question Answering for Measuring Five-Number Summary and Comparison Performance with Box Plot Yusuke Tozaki, Hisashi Miyamori, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Recently, visual question and answer (VQA) research on document and chart images, as well as natural images, has attracted much attention.In particular, there have been many studies on chart images that visualize quantities or proportions, such as bar charts, pie charts, and line graphs.However, chart images that visualize data variability, such as histograms and box plots (box-and-whisker), have not received much attention.In this paper, we propose a VQA task BoxPlotQA for box plot images and construct a new benchmark dataset for the task.Box plot images differ from other chart image VQAs in that they allow us to measure the ability to accurately read visual features such as value, width, and symmetry, and to compare these features across multiple data sets.In our experiments, we will test the effect on performance of several baseline models with and without training on the BoxPlotQA dataset, as well as the performance of different question types on real-world observed data.This study is expected to facilitate the analysis and improvement of the ability of visual language models to accurately read chart images. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Box Plot / Chart Question Answering / Visual Question Answering / Vision Language Model |
| Paper # | DE2023-16 |
| Date of Issue | 2023-09-14 (DE) |
| Conference Information | |
| Committee | DE / IPSJ-DBS / IPSJ-IFAT |
|---|---|
| Conference Date | 2023/9/21(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Kitakyushu International Conference Center |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Bigdata management, information retrieval, knowledge discovery, etc. |
| Chair | Masashi Toyoda(Univ. of Tokyo) |
| Vice Chair | Kosuke Takano(Kanagawa Inst. of Tech.) / Chiemi Watanabe(Tsukuba Univ. of Technology) |
| Secretary | Kosuke Takano(Univ. of Tsukuba) / Chiemi Watanabe(Komazawa Univ.) |
| Assistant | Takahiro Komamizu(Nagoya Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Data Engineering / Special Interest Group on Database System / Special Interest Group on Information Fundamentals and Access Technologies |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | BoxPlotQA: Visual Question Answering for Measuring Five-Number Summary and Comparison Performance with Box Plot |
| Sub Title (in English) | |
| Keyword(1) | Box Plot |
| Keyword(2) | Chart Question Answering |
| Keyword(3) | Visual Question Answering |
| Keyword(4) | Vision Language Model |
| 1st Author's Name | Yusuke Tozaki |
| 1st Author's Affiliation | Kyoto Sangyo University(Kyoto Sangyo Univ.) |
| 2nd Author's Name | Hisashi Miyamori |
| 2nd Author's Affiliation | Kyoto Sangyo University(Kyoto Sangyo Univ.) |
| Date | 2023-09-21 |
| Paper # | DE2023-16 |
| Volume (vol) | vol.123 |
| Number (no) | DE-192 |
| Page | pp.pp.31-36(DE), |
| #Pages | 6 |
| Date of Issue | 2023-09-14 (DE) |