Difference between revisions of "IAPR TC11 Newsletter 2024 03"

From TC11
Jump to: navigation, search
(Created page with "<center> 100px 140px </center> == March, 2024 == <hr> <br> Click on the buttons below to view sections of the ne...")
 
(No difference)

Latest revision as of 15:35, 28 March 2024

IAPR logo.gif Tc-11 Logo v2 72dpi.png

March, 2024



Click on the buttons below to view sections of the newsletter.

  • Message from the Editor
  • Dates and Deadlines
    • Deadlines
    • Upcoming Conferences and Events
  • Open Call for Organizing DAR Events *(repost - updated)*
  • TPDL 2024 - Call for Papers
  • DAS 2024 - Call for Papers
  • PhD Position in Machine Learning (Document Image Analysis) in Fribourg, Switzerland
  • ICDAR 2024 Competition Calls
    • ICDAR 2024 Competition: Map Text Detection, Recognition, and Linking *(updated)*
    • ICDAR 2024 Competition: Handwriting Recognition of Historical Ciphers
    • ICDAR 2024 Competition: Recognition and VQA on Handwritten Documents
    • ICDAR 2024 Competition: Reading Documents Through Aria Glasses
  • Datasets
    • TC11 Datasets Repository
      • Recently published datasets
      • Where to share datasets



Dear TC11 members,

The TC11 Datasets Repository now has three more datasets that will hopefully be beneficial to our DAR community. Among those datasets, the famous RIMES dataset released publicly by the Mitek Systems team.

Some of ICDAR 2024 competitions are published in this issue, please note the updated submission deadlines of these competitions. The CFP for DAS workshop (a satellite event of ICDAR 2024) is among the calls posted in this newsletter issue. We also re-post here the open call for organizing our major DAR events that will be held in the coming years.

I would like to highlight a CFP for the international conference on Theory and Practice of Digital Libraries (TPDL 2024) which – this year – accentuates the topic of “Document Analysis and Recognition” as one of its main themes. Other TPDL topics include “Digital Humanities and Cultural Heritage” and “Information Retrieval”.

A new job offer is announced for a PhD position in Machine Learning (Document Image Analysis) in Fribourg, Switzerland.

Finally, please have a look at the “Dates and Deadlines” section to see the updated deadlines.

Nibal Nayef, TC11 Communication Officer
( n.nayef@gmail.com )

Join us! If you are not already a member of the TC11 community, please consider joining the TC11 mailing list. Follow us on X (Twitter) (iapr_tc11): https://twitter.com/iapr_tc11



Deadlines

2024

  • March 31 Workshop proposal submission for ICPR 2024
  • March 31 Competition proposal submission for ICPR 2024
  • April 5 Proposal submission for hosting SSDA 2025
  • April 10 Paper submission for ICPR 2024 (extended)
  • April 29 Full paper submission for DAS 2024
  • April 30 Full paper Submission for TPDL 2024
  • May 7 Short paper Submission for TPDL 2024
  • June 3 Short Papers/Demos submission for DAS 2024
  • June 15 Tutorial submission for ICPR 2024

Upcoming Conferences and Events

2024

  • ICPRAI 2024. Jeju Island, South Korea (June 18-21, 2024)
  • TPDL 2024. Ljubljana, Slovenia (September 24-27, 2024)
  • ICDAR 2024. Athens, Greece (August 30 - September 4, 2024)
  • ICPR 2024. Kolkata, India (December 01-05, 2024)



The IAPR technical committees on graphics recognition (TC10) and reading systems (TC11) are regularly organizing scientific events for the Document Analysis and Recognition (DAR) community, including the ICDAR flagship conference.

In addition to specific calls for bids to host one of the events, we encourage teams to announce their interest in organizing one of the following events:

  • ICDAR: International Conference on Document Analysis and Recognition (annually; next possibility in 2027)
  • DAS: International Workshop on Document Analysis Systems (satellite event of ICDAR in even years; next possibility in 2026)
  • GREC: International Workshop on Graphics Recognition (satellite event of ICDAR in odd years; next possibility in 2025)
  • SSDA: Summer School on Document Analysis (biannually in odd years; next possibility in 2025)

You may find the details of each specific call in the previous or future issues of this newsletter or in the respective websites of each event.

Anyone interested in hosting one of these events is invited to announce their interest via email to jean-christophe.burie@univ-lr.fr and andreas.fischer@unifr.ch, in order to receive feedback and support for preparing a proposal.

Jean-Christophe Burie (Chair, TC10)
Andreas Fischer (Chair, TC11)



This is a call for research papers for the 28th International Conference on Theory and Practice of Digital Libraries (TPDL), which will be held in Ljubljana, Slovenia between 24-27 September 2024.

Website: https://tpdl2024.nuk.si/
Submission link: https://cmt3.research.microsoft.com/TPDL2024

TPDL is an international forum focused on digital libraries and their associated technical, practical, and social issues. The conference encompasses the many meanings of the term “digital libraries,” embracing the whole spectrum of the LAM (Library, Archive, and Museum) community; operational information systems with all manner of digital content; new means of selecting, collecting, organizing, and distributing digital content; and theoretical models of information media, including document genres and electronic publishing.

In 2024, TPDL is expanding its scope to prominently include Document Analysis/Recognition and Information Retrieval, acknowledging the vital role of those research areas in the creation (by means of digitization and information extraction from heterogeneous sources), access, discovery, and dissemination of digital content. This includes exploring innovative approaches to document image analysis and recognition, search algorithms, data retrieval, user engagement, and personalized content delivery within digital libraries, making these two areas central themes for this year’s conference.

Participants. Representatives from academia, cultural heritage institutions, government, industry, research communities, research infrastructures, and others are invited to participate in this annual conference. The conference draws from various research areas including computer science, information science, data science, librarianship, archival science and practice, museum studies and practice, technology, social sciences, cultural heritage, digital humanities, and the scientific communities.

Location. TPDL 2024 is hosted by the National and University Library of Slovenia and will take place in Ljubljana, Slovenia, from 24 to 27 September 2024. This is an in-person event. This choice does not exclude the possibility of following talks online, but authors of accepted papers are strongly encouraged to come and present in person. We aim to encourage discussion formally after a paper presentation and informally during social events and coffee breaks.

A selection of the best papers will be invited to submit an extended version to the International Journal of Digital Libraries.

Important Dates
Note that all deadlines are 23:59 (11:59 pm) in the AoE (Anywhere on Earth) time zone on the date specified.

  • Submission deadline for Full Papers: 30 April 2024
  • Submission deadline for Short Papers: 7 May 2024
  • Notification for Full Papers: 12 July 2024
  • Camera-ready submission: 26 July 2024
  • Conference: 24-27 September 2024

Topics in 2024 include but are not limited to, theories, models, standards, tools, and applications on the following themes: * Publishing Science: FAIR Data and Software, Research Objects, Nanopublications, Digital Preservation and Curation, Supporting Science, Reproducibility, Metadata Standards, Research Data Management, Research Output Management.

  • Data Management and Discovery: Data Search and Discovery in Digital Libraries, Data Citation and Credit Distribution, Data Lifecycle Management (creation, store, share, reuse), Data Provenance and Documentation, Linked Data and Open Data Platforms, Data Repositories and Archives, Data Stewardship and Governance, Data Integration and Harmonization.
  • Monitoring and Assessment of Science: Science of Science Studies, Scientometrics and Bibliometrics, Scholarly Communication Knowledge Graphs, Impact Analysis in Scientific Research.
  • Knowledge Creation and Dissemination: AI and Machine Learning Applications in Digital Libraries, Knowledge Bases and Organizational Systems, Entity Extraction and Semantic Linking, Ontology Development and Usage, Data as collection.
  • Digital Humanities and Cultural Heritage: Digital Cultural Heritage Preservation, Computational Linguistics and Digital Terminology, Digital Methods in History and Archaeology, Knowledge Organization in the Humanities, Digital Interfaces for Humanities Research, Digital Repatriation, Indigenous data, governance, and sovereignty.
  • Human-Computer Interaction in Digital Libraries: Design and Evaluation of User Interfaces, User Experience and Participation in Digital Libraries, Information Visualization and Analytics, Interaction Design for Diverse User Groups.
  • Information Retrieval: Advanced Search Algorithms and Techniques, Semantic Search and Indexing, User-Centric Information Retrieval Design, Information Retrieval System Evaluation, Multimodal and Multilingual Information Access, Information Behavior and Interaction in Digital Libraries.
  • Document Analysis and Recognition: Document Image Analysis and Recognition for Digital Libraries, Physical and Logical Layout Analysis, Historical Document Analysis, Document Semantics Extraction, Natural Language Processing for Document Analysis.

Contribution Types
Research Papers (12 - 15 pages + unlimited references) present high-quality, original research of relevance to the TPDL community. Submissions should detail their methods and techniques sufficiently to enable replication and reuse. Accepted papers will be published in the conference proceedings and presented as long conference talks.

Practitioner Papers (12 - 15 pages + unlimited references) present high-quality applied work of relevance to the TPDL community. Submissions should focus on results of direct relevance to practitioners and institutions in the TPDL community. Methods, tools, and techniques should be detailed sufficiently to enable application by other institutions. Accepted papers will be published in the conference proceedings and presented as long conference talks.

Short and Prototype Papers (6 - 8 pages + unlimited references) present high-quality, original research or tools or applications that are of relevance to the TPDL community. Submissions should present more focused or smaller studies, for example, preliminary results, ongoing work, or late-breaking results. Prototypes should ideally include a link to where the tool or application is available. Accepted papers will be published in the conference proceedings. Short papers are presented as short conference talks, while prototype papers are presented with a poster and a demonstration.

Submission Guidelines
Submissions should detail their methods and techniques sufficiently to enable replication and reuse. Accepted papers will be published in the conference proceedings and presented as long conference talks.
Papers must be in the Springer LNCS style:
https://www.springer.com/gp/computer-science/lncs/conference-proceedings-guidelines
Submission link: https://cmt3.research.microsoft.com/TPDL2024

Program chairs
Apostolos Antonacopoulos, University of Salford, United Kingdom
Annika Hinze, University of Waikato, New Zealand
Benjamin Piwowarski, CNRS / Sorbonne Université, France



Document Analysis System (DAS) will be organized as a satellite workshop in conjunction with ICDAR 2024. Join us at DAS 2024 to explore the cutting-edge of document analysis technologies! This workshop is an unparalleled opportunity for researchers, practitioners, and technology enthusiasts to engage in a forward-looking discourse on document analysis systems’ evolution and future directions.

We welcome papers and demos in the many categories listed here, according to Springer formatting guidelines:

  • Full papers (12-15 pages) for comprehensive research findings
  • Short papers (6-8 pages) for research in progress, and novel ideas
  • Demos. Must be accompanied by a short paper (6-8 pages)
  • Demos for Accepted ICDAR papers (4-8 pages) highlighting the application and technical detail of the accepted paper

Important Dates

  • Full Paper submission: April 29, 2024
  • Full Paper Acceptance notification: May 27, 2024
  • Full Papers Camera-ready: June 07, 2024
  • Short Papers/Demos Submission: June 03, 2024
  • Short Papers/Demos Acceptance Notification: June 10, 2024
  • Short Papers Camera-ready: June 17, 2024

For more details, please have a look at the following link:
https://das2024.seecs.edu.pk/submission.html

We are looking forward to your excellent contributions!

DAS Organizing Committee



Topic: Integration of Visual Aspects of Documents into Large Language Models
Placement: Institute of Artificial Intelligence and Complex Systems (iCoSys), HEIA Fribourg, Switzerland
Starting date: June 1, 2024 (or by arrangement)
Job link: https://icosys.ch/phd-position-in-machine-learning-document-image-analysis
Application deadline: April 15, 2024

Project description:
Large language models (LLMs) have a high potential for analyzing, recognizing, and validating scanned documents. However, they are mainly focused on the OCR text and do not take into account visual aspects, such as layout, illustrations, etc. that are of fundamental importance for document understanding.

The successful candidate will perform basic research and develop novel methods for efficient integration of visual aspects into LLMs for document understanding. A particular focus will be to obtain explainable results with respect to both visual and textual contents of the documents.

Your profile:

  • Master of Science in Computer Science (or almost finished)
  • Strong background in machine learning
  • Very good programming skills
  • Very good oral and written communication skills in English; French and/or German are a plus but not a requirement

Application:
Please provide in a single PDF file the following items:

  • Motivation letter (max 1 page)
  • CV
  • Contact information of 1 reference person (possibility to include a letter of reference)
  • Download link to your Master thesis (can be a draft version if not finished yet)

Submit your application to Prof. Dr. Andreas Fischer: andreas.fischer@hefr.ch



In the following subsections we post calls for some of ICDAR 2024 competitions. Please note the updated submission deadlines for these competitions. We encourage our DAR researchers from both academia and industry to have a look at all of this year’s ICDAR competitions.

ICDAR 2024 Competition: Map Text Detection, Recognition, and Linking (updated)

The ICDAR 2024 Competition on Map Text Detection, Recognition, and Linking is now live! If you are interested in text detection and recognition, we would like to invite you to register and participate at:
https://rrc.cvc.uab.es/?ch=28

This competition aims to address the unique challenges of detecting and recognizing textual information (e.g., place names) and linking words to form location phrases from scanned historical map images.

Important Dates

  • 1 February 2024: Training and validation data released
  • 1 March 2024: Competition test data released
  • 15 April 2024: Final results submission deadline (extended, now accepting submissions on test set)

Task Overview
The competition encompasses four tasks:

  1. word detection
  2. phrase detection
  3. word detection and recognition
  4. phrase detection and recognition. Detailed descriptions of each task can be found on the official web page.

Training, validation and test sets are already available, as well as evaluation tools. Follow us on X (Twitter) (@ICDAR24_MapText) to stay updated with the latest news!

This competition is organized by experienced competition organizers with the following affiliations: University of Minnesota (USA), Grinnell College (USA), National Geographic Agency of France, EHESS (France), EPITA (France).

Looking forward to benchmarking your innovative solutions!

— ICDAR 2024 MapText Organizers:

Contact: icdar24-maptext-contact (at) googlegroups.com
Follow us: @ICDAR24_MapText

ICDAR 2024 Competition: Handwriting Recognition of Historical Ciphers

This is a call for participation for the ICDAR 2024 Competition on Handwriting Recognition of Historical Ciphers.

https://rrc.cvc.uab.es/?ch=27&com=introduction

Handwritten Text Recognition (HTR) in low resource scenarios (i.e. when the amount of labeled data is scarce) is a challenging problem. This is particularly the case of historical encrypted manuscripts, so called ciphers, which contain secret messages, and were typically used in military or diplomatic correspondence, records of secret societies, or private letters.

In order to hide their contents, the sender and receiver created their own secret method of writing. The cipher alphabets oftentimes include digits, Latin or Greek letters, Zodiac and alchemical signs combined with various diacritics, as well as invented ones. The first step in the decryption process is the transcription of these manuscripts which is not easy due to the great variation of hand-writing styles, and cipher alphabets with a few number of pages. Although different strategies can be considered to deal with the insufficient amount of training data (e.g. few-shot learning, self-supervised learning) the performance of available HTR models is not yet satisfactory. Thus, we believe that a competition with a large number of symbol sets and scribes can boost the research of HTR in low resource scenarios.

Important Dates

  • 18 February 2024: Test data released
  • 10 May 2024: Submission of results (extended deadline)

Find more information regarding the competition, its tasks and participation guidelines at:
https://rrc.cvc.uab.es/?ch=27&com=tasks

ICDAR 2024 Competition: Recognition and VQA on Handwritten Documents

This is a call for participation in the ICDAR 2024 Competition on Recognition and VQA on Handwritten Documents. Please check the updated deadlines at the end of this call.

https://ilocr.iiit.ac.in/icdar_2024_hwd/

Handwritten OCRs have become increasingly practical with the availability of various commercial APIs, solutions, and applications. However, their widespread use is predominantly limited to English or Latin scripts, with only a few exceptions in specific Asian scripts. The challenge lies in the scarcity of datasets and the sharing of knowledge to extend these capabilities to additional languages, particularly Indian languages. Even for English, publicly accessible datasets are limited, causing academic research to fall behind advancements in industrial solutions. While achieving comprehensive recognition and understanding remains challenging, contemporary document understanding systems have evolved to address user needs through Question Answering (QA) tasks. This approach has also been applied to Document Question Answering (DocVQA). The current juncture presents an opportunity to explore information extraction from handwritten documents, where the complexities of recognition and layout (both structure and content) pose ongoing challenges.

Like previous competitions, ICFHR 2022 IHTR and ICDAR 2023 IHTR, the proposed competition continues this effort with even more unique datasets. It introduces two novel tasks: page level recognition and reading, and visual question answers on handwritten documents. Our challenge is centered around handwritten document recognition. This competition stands as a dynamic catalyst, igniting the passion and creativity of researchers to pioneer ground-breaking solutions in the realm of handwritten document analysis. By providing a platform for innovation and algorithm design, it serves as a driving force, inspiring participants to push the boundaries of what is achievable in understanding and interpreting handwritten documents.

Important Dates

  • Registration Opens : January 20, 2024
  • Training Data Release : February 29, 2024
  • Updated Training Data : April 15, 2024 (updated)
  • Test Data Release : May 1, 2024 (updated)
  • Registration Close : May 10, 2024 (updated)
  • Results Submission Deadline : May 15, 2024 (updated)
  • Winner Announcement : May 25, 2024 (updated)

Find more information regarding the competition and its tasks at:
https://ilocr.iiit.ac.in/icdar_2024_hwd/task.html

ICDAR 2024 Competition: Reading Documents Through Aria Glasses

This is a call for participation in the ICDAR 2024 Competition on Reading Documents Through Aria Glasses. Please check the updated deadlines at the end of this call.

https://ilocr.iiit.ac.in/icdar_2024_rdtag/

Understanding text within everyday documents is a crucial aspect of human communication and knowledge acquisition. However, while there has been tremendous progress in optical character recognition (OCR), it is still a challenge to recognize the content of documents from egocentric, first-person devices, such as AR headsets and smart glasses. This is due to the unique sensor configurations of egocentric devices, and the motion characteristics of data captured. We believe enabling computers to read and understand text from egocentric devices could unlock the next generation of contextual AI systems, by enabling devices to learn and interpret information through casual, lifelong activities, just like humans do.

To accelerate research in this area, we are pleased to announce ‘Reading documents through Aria Glasses’, a new research challenge in collaboration with Project Aria from Meta Reality Labs Research. The competition harnesses data captured from Project Aria glasses, to consider the unique challenges posed by egocentric devices, including power and sensor constraints, lighting conditions, and reading positions from everyday scenes.

To tackle a range of challenges, we introduce three sub-tasks:

    1. Isolated Word Recognition in Low Resolution (Task A)
    1. Prediction of Reading Order (Task B)
    1. Page Level Recognition (Task C)

Important Dates

  • Registration Opens : January 20, 2024
  • Training Data Release : February 29, 2024
  • Updated Training Data : April 15, 2024 (updated)
  • Test Data Release : May 1, 2024 (updated)
  • Registration Close : May 10, 2024 (updated)
  • Results Submission Deadline : May 15, 2024 (updated)
  • Winner Announcement : May 25, 2024 (updated)

Find more information regarding the competition and its tasks at:
https://ilocr.iiit.ac.in/icdar_2024_rdtag/task.html



TC11 Datasets Repository

Recently published datasets

The famous RIMES dataset has finally been released publicly by the Mitek Systems team. French DAR researchers in particular appreciate this release as it is a valuable effort from a private company.

RIMES datasets collection

  • Contributed: 2024-03-14
  • Keywords: OCR, Handwritten Text Recognition, RIMES, Document Image Analysis, Information Extraction, Writer recognition, Document Structure Identification
  • URL: https://zenodo.org/records/10812725

Here are more datasets for our DAR community:

Handwritings Datasets for Urdu & English (IPC-WritDAR)

Urdu Artificial Text Dataset (IPC-ArtifDAR)

Where to share datasets

Did you know it? We have two official places for datasets:
- Our historical platform for storage and listing: http://datasets.iapr-tc11.org

TC11 maintains a collection of datasets that can be found online in the TC11 Datasets Repository.

If you have new datasets (e.g., from competitions) that you wish to share with the research community, please use the online upload form. For questions and support, please contact the TC11 Dataset Curator (contact information is below).

Joseph Chazalon (TC11 Dataset Curator)
( joseph.chazalon@lrde.epita.fr )


Call for Contributions: To contribute news items, please send a short email to the editor, [Nibal Nayef](mailto:n.nayef@gmail.com). Contributions might include conference and workshop announcements/updates/reports, career opportunities, book reviews, or anything else of interest to the TC-11 community.

Subscription: This newsletter is sent to subscribers of the IAPR TC11 mailing list. To join the TC-11 mailing list, please click on this link: Join the TC-11 Mailing List. To manage your subscription, please visit the mailing list homepage: TC-11 Mailing List Homepage.