DAS 2022, the 15th IAPR International Workshop on Document Analysis System, received 88 full-paper submissions. After careful and intensive review, for which we thank the reviewers, 31 papers are accepted for oral presentation and 21 for poster presentation.

(The list is sorted by ORAL/POSTER and EasyChair submission number.)

#AuthorsTitlePresentation type
2Rafael Lins, Rodrigo B. Bernardino, Ricardo Barboza and Raimundo OliveiraThe Winner Takes It All: choosing the “best” binarization algorithm for photographed documentsOral
6Shuang Liu, Renshen Wang, Michalis Raptis and Yasuhisa FujiiUnified Line and Paragraph Detection by Graph Convolutional NetworksOral
9Christoph Wick, Jochen Zöllner and Tobias GrüningRescoring Sequence-to-Sequence Models for Text Line Recognition with CTC-PrefixesOral
12Michael Koepf, Florian Kleber and Robert SablatnigWriter Identification and Writer Retrieval using Vision Transformer for Forensic DocumentsOral
14Hussein Mohammed, Agnieszka Helman-Wazny, Claudia Colini, Wiebke Beyer and Sebastian BoschPattern Analysis Software Tools (PAST) for Ancient Written ArtefactsOral
16Martin Maarand, Yngvil Beyer, Andre Kåsen, Knut Fosseide and Christopher KermorvantA Comprehensive Comparison of Open-Source Libraries for Handwritten Text Recognition in NorwegianOral
19Christian Reul, Stefan Tomasek, Florian Langhanki and Uwe SpringmannOpen Source Handwritten Text Recognition on Medieval Manuscripts using Mixed Models and Document-Specific FinetuningOral
20Hadia Showkat Kawoosa, Mandhatya Singh, Manoj Manikrao Joshi and Puneet GoyalNCERT5K-IITRPR: A Benchmark Dataset for Non-Textual Component Detection in School BooksOral
23Claire Bizon Monroc, Blanche Miret, Marie-Laurence Bonhomme and Christopher KermorvantA Comprehensive Study of Open-source Libraries for Named Entity Recognition on Handwritten Historical DocumentsOral
26Oliver Tüselmann and Gernot FinkNamed Entity Linking on Handwritten Document ImagesOral
27José Andrés, Alejandro H. Toselli and Enrique VidalApproximate Search for Keywords in Handwritten Text ImagesOral
28Killian Barrere, Yann Soullard, Aurélie Lemaitre and Bertrand CoüasnonA Light Transformer-Based Architecture for Handwritten Text RecognitionOral
33Ladislav Lenc, Jiří Martínek, Martin Prantl, Josef Baloun and Pavel KrálHistorical Map Toponym Extraction for Efficient Information RetrievalOral
36Giorgos Sfikas, George Retsinas, Angelos P. Giotis, Basilis Gatos and Christophoros NikouKeyword Spotting with Quaternionic ResNet: Application to Spotting in Greek ManuscriptsOral
37Martin Kišš, Jan Kohút, Karel Beneš and Michal HradišImportance of Textlines in Historical Document ClassificationOral
46Simon Schiff, Sylvia Melzer, Eva Wilden and Ralf MöllerTEI-based Interactive Critical EditionsOral
48Josep Brugués i Pujolràs, Lluis Gomez and Dimosthenis KaratzasA Multilingual Approach to Scene Text Visual Question AnsweringOral
54Thomas Constum, Nicolas Kempf, Thierry Paquet, Pierrick Tranouez and Clément ChatelainRecognition and information extraction in historical handwritten tables: toward understanding early 20th century Paris censusOral
56José Andrés, Jose Ramón Prieto, Emilio Granell, Verónica Romero, Joan Andreu Sánchez and Enrique VidalInformation Extraction from Handwritten Tables in Historical DocumentsOral
58Ramon Pires, Fabio Souza, Guilherme Rosa, Roberto Lotufo and Rodrigo NogueiraSequence-to-Sequence Models for Extracting Information from Registration and Legal DocumentsOral
60Nathalie Abadie, Edwin Carlinet, Joseph Chazalon and Bertrand DuménieuA Benchmark of NER Approaches in Historical DocumentsOral
63G Nagendar and Ramachandrula SitaramContrastive Graph Learning with Graph Convolutional NetworksOral
64Xiaotong Ji, Yan Zheng, Daiki Suehiro and Seiichi UchidaRevealing Reliable Signatures by Learning Top-Rank PairsOral
66Masaya Ueda, Akisato Kimura and Seiichi UchidaFont Shape-to-Impression TranslationOral
71Prabhat Kumar Bharti, Tirthankar Ghosal, Mayank Agrawal and Asif EkbalHow Confident Was Your Reviewer? Estimating Reviewer Confidence From Peer Review TextsOral
72Yusuke Nagata, Jinki Otao, Daichi Haraguchi and Seiichi UchidaTrueType Transformer: Character and Font Style Recognition in Outline FormatOral
74George Retsinas, Giorgos Sfikas, Basilis Gatos and Christophoros NikouBest Practices for a Handwritten Text Recognition systemOral
76George Retsinas, Giorgos Sfikas, Basilis Gatos and Christophoros NikouOn-The-Fly Deformations for Keyword SpottingOral
87Joan Andreu Sanchez, Enrique Vidal and Vicente BoschEffective Crowdsourcing in the EDT Project with Probabilistic IndexesOral
90Thibault Douzon, Christophe Garcia, Stefan Duffner and Jérémy EspinasImproving Information Extraction on Business Documents with Specific Pre Training TasksOral
92Raphaela Heil, Ekta Vats and Anders HastPaired Image to Image Translation for Strikethrough Removal From Handwritten WordsOral
5Gonzalo Santamaría, Cesar Dominguez, Jónathan Heras, Eloy Mata and Vico PascualCombining image processing techniques, OCR, and OMR for the digitization of musical booksPoster 
15Wassim Swaileh, Michel Jordan and Dimitrios Kotzinos3D Modelling Approach for Ancient Floor Plans’ Quick BrowsingPoster 
17Mohamed El Baha, Olivier Augereau, Sofiya Kobylyanskaya, Ioana Vasilescu and Laurence DevillersEye Got It : a System for Automatic Calculation of the Eye-Voice Span.Poster 
18David Villanova-Aparisi, Carlos-D. Martínez-Hinarejos, Verónica Romero and Moisés Pastor-GadeaEvaluation of Named Entity Recognition in handwritten documentsPoster 
21Ahmad Droby, Daria Vasyutinsky Shapira, Irina Rabaev, Berat Kurar and Jihad El-SanaHard and Soft Labeling for Hebrew Paleography: A Case StudyPoster 
22Solène Tarride, Aurélie Lemaitre, Bertrand Coüasnon and Sophie TardivelA comparative study of information extraction strategies using an attention-based neural networkPoster 
34Ibrahim Souleiman Mahamoud, Michaël Coustaty, Aurélie Joseph, Vincent Poulain d’Andecy and Jean-Marc OgierQalayout :Question answering Layout based on multimodal Attention for visual question answering on corporate DocumentPoster 
41Konstantina Nikolaidou, Richa Upadhyay, Mathias Seuret and Marcus LiwickiInvestigating the Effect of using Synthetic and Semi-synthetic Images for Historical Document ClassificationPoster 
45Adrià Molina Rodríguez, Josep Lladós Canet, Oriol Ramos Terrades and Lluis Gomez BigordaA Generic Date Estimation System for Historical Document Images based on Ordinal ClassificationPoster 
51Alexander Mattick, Martin Mayr, Andreas Maier and Vincent ChristleinIs multi-task learning always better?Poster 
55Mathieu Francois, Véronique Eglin and Maxime BiouText detection and post-OCR correction in Engineering DocumentsPoster 
59Dmitrijs Kass and Ekta VatsAttentionHTR: Handwritten Text Recognition Based on Attention Encoder-Decoder NetworksPoster 
62Sergi Garcia, George Tom, Sangeeth Battu, Minesh Mathew, Marçal Rusiñol, C.V. Jawahar and Dimosthenis KaratzasRead while you Drive – Multilingual Text Tracking on the RoadPoster 
67Athar Sefid and C. Lee GilesSciBERTSUM: Extractive Summarization for Scientific DocumentsPoster 
73Hai Thi Tuyet Nguyen, Adam Jatowt, Mickael Coustaty and Antoine DoucetReadOCR: A Novel Dataset and Readability Assessment of OCRed TextsPoster 
75Panagiotis Kaddas and Basilis GatosUsing Multi-level Segmentation Features for Document Image ClassificationPoster 
78Muhammad Atif Butt, Adnan Ul-Hasan and Faisal ShafaitTraffSign: Multilingual Traffic Signboard Text Detection and Recognition for Urdu and EnglishPoster 
79Martin Mayr, Alex Felker, Andreas Maier and Vincent ChristleinCombining Visual and Linguistic Models for a Robust Recipient Line Recognition in Historical DocumentsPoster 
80Borak Madi, Reem Alaasam, Ahmad Droby and Jihad El-SanaHST-GAN: Historical Style Transfer GAN for Generating Historical Text ImagesPoster 
89Sofiane Medjram and Véronique EglinChallenging children handwriting recognition study exploiting synthetic, mixed and real dataPoster 
98Richin Sukesh, Mathias Seuret, Anguelos Nikolaou, Martin Mayr and Vincent ChristleinA Fair Evaluation of Deep Learning-based Binarization MethodsPoster