Text region detection in historical astronomical diagrams
Not on the map. Why?
Abstract (arXiv)
Text detection is a crucial task in the analysis of historical documents. While datasets and benchmarks exist for text detection in manuscripts and maps, the study of text in mathematical diagrams has received little attention. To address this, we introduce a large-scale, diverse, open-access dataset of 948 historical astronomical diagrams containing 10,940 oriented polygonal text regions. Our dataset spans ten centuries (8th to 18th) and seven main linguistic traditions: Arabic and Persian (115), Chinese (332), Byzantine (233), Latin (185), Hebrew (48), and Sanskrit (35). It captures a wide range of diagram styles and textual content, from symbols to multi-line paragraphs. Each text instance is annotated with ordered polygons that precisely delineate text regions and encode the reading direction. In addition, we annotated the 2,293 regions in Latin diagrams with 20 class labels. We evaluated several strong baselines on our dataset, including TESTR, DeepSolo++, and Poly-DETR, a simple extension of DINO-DETR that we design to predict ordered polygon vertices. Poly-DETR achieves state-of-the-art performance on the MTHv2 and cBAD2019 benchmarks and provides a solid, simple baseline on our dataset. Code and dataset available online.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 26 of its 53 references that are in the map. First posted on arXiv 14 Jun 2026.
Builds on
26 of 53 references.
References · 26
- 2020End-to-End Object Detection with Transformerscited by 7,414
- 2017Mask R-CNNcited by 7,277
- 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
- 2022DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detectioncited by 1,319
- 2018Scene Text Detection and Recognition: The Deep Learning Eracited by 114
- 2019Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Networkcited by 114
- 2022Text Spotting Transformerscited by 70
- 2020TableNet: Deep Learning model for end-to-end Table detection and Tabular data extraction from Scanned Document Imagescited by 69
- 2018dhSegment: A generic deep-learning approach for document segmentationcited by 32
- 2018A Two-Stage Method for Text Line Detection in Historical Documentscited by 26
- 2019Omnidirectional Scene Text Detection with Sequential-free Box Discretizationcited by 26
- 2020Joint Layout Analysis, Character Detection and Recognition for Historical Document Digitizationcited by 21
14 more
- 2024The Manga Whisperer: Automatically Generating Transcriptions for Comicscited by 21
- 2020docExtractor: An off-the-shelf historical document element extractioncited by 20
- 2020HORAE: an annotated dataset of books of hourscited by 13
- 2019ICDAR 2019 Historical Document Reading Challenge on Large Structured Chinese Family Recordscited by 10
- 2021Synthetic Map Generation to Provide Unlimited Training Data for Historical Map Text Detectioncited by 10
- 2023The mapKurator System: A Complete Pipeline for Extracting and Linking Text from Historical Mapscited by 8
- 2023DeepSolo++: Let Transformer Decoder with Explicit Points Solo for Multilingual Text Spottingcited by 7
- 2025Hyper-Local Deformable Transformers for Text Spotting on Historical Mapscited by 7
- 2022Robust Text Line Detection in Historical Documents: Learning and Evaluation Methodscited by 4
- 2025LRANet++: Low-Rank Approximation Network for Accurate and Efficient Text Spottingcited by 4
- 2021Palmira: A Deep Deformable Network for Instance Segmentation of Dense and Uneven Layouts in Handwritten Manuscriptscited by 2
- 2024Historical Astronomical Diagrams Decomposition in Geometric Primitivescited by 2
- 2024SegHist: A General Segmentation-based Framework for Chinese Historical Document Text Line Detectioncited by 2
- 2018Baseline Detection in Historical Documents using Convolutional U-Netscited by 1
Built on by
Not cited yet.