Document Analysis and Recognition – ICDAR 2021 Workshops

1. Front Matter

   Pages i-xx

   PDF

2. ICDAR 2021 Workshop on Graphics Recognition (GREC)

   1. Front Matter

      Pages 1-5

      PDF

   2. Relation-Based Representation for Handwritten Mathematical Expression Recognition

      • Thanh-Nghia Truong, Huy Quang Ung, Hung Tuan Nguyen, Cuong Tuan Nguyen, Masaki Nakagawa

      Pages 7-19

   3. A Public Ground-Truth Dataset for Handwritten Circuit Diagram Images

      • Felix Thoma, Johannes Bayer, Yakun Li, Andreas Dengel

      Pages 20-27

   4. A Self-supervised Inverse Graphics Approach for Sketch Parametrization

      • Albert Suso, Pau Riba, Oriol Ramos Terrades, Josep Lladós

      Pages 28-42

   5. Border Detection for Seamless Connection of Historical Cadastral Maps

      • Ladislav Lenc, Martin Prantl, Jiří Martínek, Pavel Král

      Pages 43-58

   6. Data Augmentation for End-to-End Optical Music Recognition

      • Juan C. López-Gutiérrez, Jose J. Valero-Mas, Francisco J. Castellanos, Jorge Calvo-Zaragoza

      Pages 59-73

   7. Graph-Based Object Detection Enhancement for Symbolic Engineering Drawings

      • Syed Mizanur Rahman, Johannes Bayer, Andreas Dengel

      Pages 74-90

   8. ScanSSD-XYc: Faster Detection for Math Formulas

      • Abhisek Dey, Richard Zanibbi

      Pages 91-96

   9. Famous Companies Use More Letters in Logo: A Large-Scale Analysis of Text Area in Logo

      • Shintaro Nishi, Takeaki Kadota, Seiichi Uchida

      Pages 97-111

   10. MediTables: A New Dataset and Deep Network for Multi-category Table Localization in Medical Documents

       • Akshay Praveen Deshpande, Vaishnav Rao Potlapalli, Ravi Kiran Sarvadevabhatla

       Pages 112-124

   11. Online Analysis of Children Handwritten Words in Dictation Context

       • Omar Krichen, Simon Corbillé, Eric Anquetil, Nathalie Girard, Pauline Nerdeux

       Pages 125-140

   12. A Transcription Is All You Need: Learning to Align Through Attention

       • Pau Torras, Mohamed Ali Souibgui, Jialuo Chen, Alicia Fornés

       Pages 141-146

   13. Accurate Graphic Symbol Detection in Ancient Document Digital Reproductions

       • Zahra Ziran, Eleonora Bernasconi, Antonella Ghignoli, Francesco Leotta, Massimo Mecella

       Pages 147-162Open Access

3. ICDAR 2021 Workshop on Camera-Based Document Analysis and Recognition (CBDAR)

   1. Front Matter

      Pages 163-165

      PDF

   2. Inscription Segmentation Using Synthetic Inscription Images for Text Detection at Stone Monuments

      • Naoto Morita, Ryunosuke Inoue, Masashi Yamada, Takatoshi Naka, Atsuko Kanematsu, Shinya Miyazaki et al.

      Pages 167-181

   3. Transfer Learning for Scene Text Recognition in Indian Languages

      • Sanjana Gunna, Rohit Saluja, C. V. Jawahar

      Pages 182-197

   4. How Far Deep Learning Systems for Text Detection and Recognition in Natural Scenes are Affected by Occlusion?

      • Aline Geovanna Soares, Byron Leite Dantas Bezerra, Estanislau Baptista Lima

      Pages 198-212

   5. CATALIST: CAmera TrAnsformations for Multi-LIngual Scene Text Recognition

      • Shivam Sood, Rohit Saluja, Ganesh Ramakrishnan, Parag Chaudhuri

      Pages 213-228

   6. DDocE: Deep Document Enhancement with Multi-scale Feature Aggregation and Pixel-Wise Adjustments

      • Karina O. M. Bogdan, Guilherme A. S. Megeto, Rovilson Leal, Gustavo Souza, Augusto C. Valente, Lucas N. Kirsten

      Pages 229-244

This book constitutes the proceedings of the international workshops co-located with the 16th International Conference on Document Analysis and Recognition, ICDAR 2021, held in Lausanne, Switzerland, in September 2021.The total of 59 full and 12 short papers presented in this book were carefully selected from 96 contributions and divided into two volumes. Part I contains 29 full and 4 short papers that stem from the following meetings: ICDAR 2021 Workshop on Graphics Recognition (GREC); ICDAR 2021 Workshop on Camera-Based Document Analysis and Recognition (CBDAR); ICDAR 2021 Workshop on Arabic and Derived Script Analysis and Recognition (ASAR 2021); ICDAR 2021 Workshop on Computational Document Forensics (IWCDF). The main topics of the contributions are document processing; physical and logical layout analysis; text and symbol recognition; handwriting recognition; signature verification and document forensics, and others.

“Accurate Graphic Symbol Detection in Ancient Document Digital Reproductions” is available open access under a Creative Commons Attribution 4.0 International License via link.springer.com.

• artificial intelligence
• character recognition
• computational linguistics
• computer systems
• computer vision
• data mining
• databases
• image analysis
• image processing
• image quality
• image reconstruction
• image segmentation
• information retrieval
• machine learning
• Natural Language Processing (NLP)
• natural languages
• neural networks
• optical character recognition
• pattern recognition
• semantics

• Boise State University, Boise, USA

  Elisa H. Barney Smith

• Indian Statistical Institute, Kolkata, India

  Umapada Pal

Policies and ethics