Skip to main navigation Skip to search Skip to main content

LineFormer: Line Chart Data Extraction Using Instance Segmentation

  • SUNY Buffalo

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

12 Scopus citations

Abstract

Data extraction from line-chart images is an essential component of the automated document understanding process, as line charts are a ubiquitous data visualization format. However, the amount of visual and structural variations in multi-line graphs makes them particularly challenging for automated parsing. Existing works, however, are not robust to all these variations, either taking an all-chart unified approach or relying on auxiliary information such as legends for line data extraction. In this work, we propose LineFormer, a robust approach to line data extraction using instance segmentation. We achieve state-of-the-art performance on several benchmark synthetic and real chart datasets. Our implementation is available at https://github.com/TheJaeLal/LineFormer.

Original languageEnglish
Title of host publicationDocument Analysis and Recognition – ICDAR 2023 - 17th International Conference, Proceedings
EditorsGernot A. Fink, Rajiv Jain, Koichi Kise, Richard Zanibbi
PublisherSpringer Science and Business Media Deutschland GmbH
Pages387-400
Number of pages14
ISBN (Print)9783031417337
DOIs
StatePublished - 2023
Event2023 International Workshops co-located with the 17th International Conference on Document Analysis and Recognition, ICDAR 2023 - San José, United States
Duration: Aug 24 2023Aug 26 2023

Publication series

NameLecture Notes in Computer Science
Volume14191 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference2023 International Workshops co-located with the 17th International Conference on Document Analysis and Recognition, ICDAR 2023
Country/TerritoryUnited States
CitySan José
Period08/24/2308/26/23

Keywords

  • Chart Data Extraction
  • Chart OCR
  • Instance Segmentation
  • Line Charts

Fingerprint

Dive into the research topics of 'LineFormer: Line Chart Data Extraction Using Instance Segmentation'. Together they form a unique fingerprint.

Cite this