IAPR TC11 Newsletter 2020 12
Click on the buttons below to view sections of the newsletter.
- Message from the Editor
- Message from the Chair
- Dates and Deadlines
- Upcoming Conferences and Events
- TC10/11 Meeting
- Invitation to participate in online meeting
- Call for bids for organising DAS 2022
- Open call for expressions of interest for organising future TC10 or TC11 events
- ICDAR 2021
- ICDAR 2021: 2nd Call for Papers
- ICDAR 2021: Call for Tutorials (deadline extended)
- ICDAR 2021: Workshops
- ICDAR 2021: Long-Term Competition on Document Visual Question Answering (DocVQA)
- ICDAR 2021: Competition on Historical Map Segmentation
- TC11 Datasets Repository *(repost)*
- PostDoc-position in Computer Vision and Machine Learning for Document Image Fraud Detection
A challenging year concludes, during which the TC11 community could not meet and exchange in person as usual. Nevertheless, we are looking back to a large number of activities and initiatives during 2020, which are summarized in the message from the TC11 chair below.
During the holiday season, please consider browsing through the numerous items in this newsletter to stay up to date:
- An invitation for the annual TC10/11 meeting, which will be held online early next year. The event is open to all TC11 members receiving this newsletter.
- A call for bids to host DAS 2020.
- An open call for bids to host a future TC10/11 event.
- The second call for papers for ICDAR 2021. Note that the paper template and the number of pages has changed when compared with previous editions.
- The list of accepted ICDAR workshops.
- A deadline extension for proposing an ICDAR tutorial.
- A call for participating in the ICDAR long-term competition on Document Visual Question Answering (DocVQA).
- A call for participating in the ICDAR competition on Historical Map Segmentation (HistMapSeg).
- An open PostDoc position at L3i in La Rochelle, France.
With that I wish you happy holidays and a happy New Year! Stay safe.
Andreas Fischer, TC11 Communications Officer
( email@example.com )
As 2020 comes to an end, I wanted to wish everyone a happy new year, full of health, in a world that has hopefully come a bit closer together through our common struggle during this pandemic.
2020 has been a challenging year. Still, the TC11 community has managed to keep on track, grow and evolve. We did have to change plans and make last minute decisions based on inadequate data, but after all I think we have achieved all we had planned for this year and more.
For the first time in their long history, DAS and ICFHR were held virtually. Instead of welcoming everyone to Wuhan and to Dortmund, the organisers brought Wuhan and Dortmund to the home offices and living rooms of the online participants. In a sense, these were the first worldwide editions of DAS and ICFHR, allowing anyone with an interest in our field and an internet connection to participate. This was reflected in the participation patterns, DAS had the greatest participation ever, reaching 160 participants, and ICFHR reached 124. Many of the participants attended for the first time – welcome! And according to the post-event questionnaires, pretty much everyone agreed that both events were of high scientific quality, and a rewarding experience.
I want to wholeheartedly thank Cheng-Lin, Shijian, Jean-Marc, Xiang, Gernot, Rejean, Lambert and everyone on the organisation teams of DAS 2020 and ICFHR 2020 for their dedication and for managing to make the most out of an admittedly difficult situation.
In parallel, during 2020 we have advanced a lot in the definition of a strategic plan for the document analysis and recognition community. As we promised to do during the TC10/11 meeting at ICDAR 2019, we have set up a Task Force to work on a mid-long term strategy for the community, and we actually managed to do a first face-to-face meeting back in January – just before the world closed down! Since then we have been working online and we gave short updates on our work in DAS and ICFHR this year. You can watch the recording of the DAS session here:
Those of you who were in ICDAR 2017, ICDAR 2019, and the corresponding “Future of DAR” workshops have had first-hand participation in numerous discussions and hands-on activities that informed this strategy; and those of you who did not, do not worry, you will still have chances to do so. Watch this space, we will be circulating the current status of the “DAR strategic plan” soon and will be asking for your feedback!
Next stop, ICPR and then ICDAR 2021! See you all there! In the meantime, have a relaxing holiday, and all the best for the new year!
Dr Dimosthenis Karatzas, TC11 Chair
Upcoming Conferences and Events
- ICPR 2020. Virtual (January 10-15)
- Online TC10/11 meeting (January 20)
- ICDAR 2021. Lausanne, Switzerland (September 5-10)
2022 and Later
- ICFHR 2022. Hyderabad, India (December, 2022)
Invitation to participate in online meeting
You are all invited to attend the next TC10/11 meeting that will take place online on Wednesday, January 20, 12:00 – 13:00 CET. The meeting is open to all TC11 members receiving this newsletter.
Link for connecting to the online meeting: TC10/11 Meeting
Note that it is habitual to host a joint TC10/11 meeting during the ICPR. Since ICPR will take place as a virtual conference this year, we have decided to decouple the TC10/11 meeting from the conference and hold it separately the week after ICPR as a virtual meeting. This way the TC10/11 meeting will not be bound to the timing of ICPR coffee breaks, while it will be open to everyone.
Call for bids for organising DAS 2022
Following the successful organisation of the 14th IAPR International Workshop on Document Analysis Systems, hosted virtually from Wuhan (China) last July, we are now soliciting proposals for organising and hosting DAS 2022.
The DAS workshop has become one of the signature events for TC11. DAS 2022 will build on the tradition established by past DAS workshops held in Kaiserslautern, Germany (1994), Malvern, PA (1996), Nagano, Japan (1998), Rio de Janeiro, Brazil (2000), Princeton, NJ (2002), Florence, Italy (2004), Nelson, New Zealand (2006), Nara, Japan (2008), Boston, MA (2010), and Gold Coast, Australia (2012), Tours-Loire Valley, France (2014), Santorini, Greece (2016), Vienna, Austria (2018), and Wuhan, China (2020).
Individuals and groups who are interested in Document Analysis Systems are invited to submit proposals for organizing and hosting DAS 2022. The event should preferably take place in late summer/fall but is not limited to this period.
The submission deadline is March 31st, 2021. Proposals should be submitted to the TC11 chair (Dimosthenis Karatzas) and vice-chair (Gernot Fink).
If you have a potential interest in preparing a proposal, please let us know in advance. Such an expression of interest is not a commitment to make a formal proposal nor an official bid, but it would help us monitor the bid preparation activity and will give us a chance to help with the preparation of the bids.
Similarly, if you need further information concerning DAS, please feel free to contact us.
The final selection among competing proposals will be made short after the deadline by the DAS Steering Committee, which is composed of all those who have themselves organized or contributed substantially to past DAS workshops.
Dr Dimosthenis Karatzas, TC11 Chair (firstname.lastname@example.org)
Dr Gernot Fink, TC11 Vice-Chair (email@example.com)
Open call for expressions of interest for organising future TC10 or TC11 events
Due to the current pandemic situation, and understanding the difficulties in putting together a complete bid under the current circumstances, the TC10 and TC11 teams have opted not to open specific calls for organising any future events planned for after 2023.
Nevertheless, we would like to probe the community, and invite our colleagues to start thinking about whether they would like to potentially organise a future TC10 or TC11 event (e.g. conference, workshop, summer school).
We will be gathering expressions of interest for a few months. These are not meant to be full proposals, but rather an informal message to the TC10/TC11 chair, indicating a possible interest to organise one of the future events.
If you think you can take on the organisation of such a future event, please send us a quick message, indicating if possible where, when and which event you would be interested in organising.
Dr Dimosthenis Karatzas, TC11 Chair (firstname.lastname@example.org)
Dr Alicia Fornes, TC10 Chair (email@example.com)
ICDAR 2021: 2nd Call for Papers
The 16th International Conference on Document Analysis and Recognition
September 5-10, 2021 - Lausanne, Switzerland
ICDAR is the premier international event for scientists and practitioners involved with document analysis and recognition, a field of growing importance in the current age of digital transition. In 2021, the 16th edition of this flagship conference will be held in Switzerland for the first time.
The organizers are committed to holding a safe and secure ICDAR 2021 in the face of the worldwide COVID-19 health crisis. In the event it becomes necessary, ICDAR 2021 will be held with options for remote participation. All accepted papers will be presented at the conference and published in the ICDAR 2021 proceedings, whether or not the conference takes place physically, virtually, or in a hybrid format.
08 Feb 2021: Paper submission deadline 26 Apr 2021: Notification of acceptance 17 May 2021: Camera-ready version submission deadline 05-07 Sep 2021: Workshops, tutorials, and doctoral consortium 08-10 Sep 2021: Main conference dates
Topics of interest to the conference include, but are not limited to:
- Document image processing
- Physical and logical layout analysis
- Text and symbol recognition
- Handwriting recognition
- Document analysis systems
- Document classification
- Indexing and retrieval of documents
- Document synthesis
- Extracting document semantics
- NLP for document understanding
- Document summarization and translation
- Office automation
- Human document interaction
- Multimedia document analysis
- Mobile text recognition
- Pen‐based document analysis
- Scene text detection and recognition
- Graphics recognition
- Recognition of tables and formulas
- Historical document analysis
- Signature verification
- Document forensics and provenance analysis
- Medical document analysis
- Document analysis for social good
- Document analysis for literature search
- Gold-standard benchmarks and data sets
Submission and Review
ICDAR 2021 will follow a double blind review process. Authors should not include their names and affiliations anywhere in the manuscript. Authors should also ensure that their identity is not revealed indirectly by citing their previous work in the third person, and omit acknowledgements until the camera-ready version.
Springer Lecture Notes in Computer Science is the Publisher for ICDAR 2021
Recent past ICDAR proceedings have been published using IEEE Computer Society Press. Building on successful collaborations between IAPR and Springer, the ICDAR advisory board has decided to publish the ICDAR 2021 proceedings under the Springer Lecture Notes in Computer Science (LNCS) series. Click here for more details.
This change provides the proceedings of the conference and the workshops with an excellent online accessibility, including free access to SpringerLink via links on the conference website during one year after the publication and free access for everyone in SpringerLink four years after the publication. In addition, the conference benefits from lower publishing costs.
This change also affects the paper format and length. Detailed instructions are provided below.
The EasyChair paper submission link will be available soon on the conference webpage.
Paper Format and Length
Papers accepted for the conference will be allocated up to 15 pages in the proceedings. Submissions are expected to be in the range of 12-15 pages.
No-show papers will be removed from the online conference proceedings and will not be indexed. No-show papers that were not withdrawn before the Technical Program was announced will be identified as “No-Show” on the conference website.
At least one author of each accepted paper MUST register the paper at full-rate (i.e., a student registration will not suffice). Papers without a confirmed registration will be withdrawn by the conference organizers.
By submitting a manuscript to ICDAR 2021, authors acknowledge that it has not been previously published or accepted for publication in substantially similar form in any peer-reviewed venue including journals, conferences, or workshops. Furthermore, authors agree that no paper substantially similar in content has been or will be submitted to a journal, or another conference or workshop during the review period (February 8, 2021 – April 26, 2021). Note that this restriction does not apply to papers issued as unpublished technical reports or included in self-archive repositories (departmental, arXiv.org, etc.) that are not peer-reviewed.
Note: authors who submitted a paper to the IJDAR-ICDAR journal track via the IJDAR website should NOT re-submit the same paper to the ICDAR conference. For questions, please contact the ICDAR 2021 program committee chairs (firstname.lastname@example.org).
IAPR Ethical Requirements for Authors
By submitting a manuscript to ICDAR 2021, authors agree to adhere the IAPR Statement of Ethics. The IAPR requires that all authors wishing to present a paper declare that:
- The paper is substantially original and that no paper substantially similar in content has been submitted or will be submitted to any other conference or journal during the ICDAR 2021 review period.
- The paper reflects the work of the authors and contains no plagiarism, and
- If accepted, the paper will be presented by one of the authors in person.
The IAPR retains the right to withdraw any paper found in violation of these requirements, and to exclude the authors in question from future IAPR community activities.
The full IAPR Statement of Ethics can be found here.
Automated Plagiarism Detection
Submissions may be automatically screened for signs of plagiarism, including self-plagiarism. Please use extra care that you have not copied verbatim blocks of text from any other published sources, including your own.
Author Response Phase Instructions
The author response phase which takes place after the first round of reviews is optional and meant to provide authors with an opportunity to correct factual errors in a review, or to supply additional information as requested by reviewers. It is NOT intended as a way to add new contributions (theorems, algorithms, experiments) or as an opportunity to argue with the opinions of reviewers.
Andreas Fischer, Rolf Ingold, and Marcus Liwicki - ICDAR 2021 General Chairs
ICDAR 2021: Call for Tutorials (deadline extended)
Jan 11, 2020 (extended) Proposal Due Jan 31, 2020 (extended) Acceptance Notification Sep 5-7, 2021 Dates of Tutorials
The ICDAR 2021 Organizing Committee invites proposals for tutorials that will be held on September 5-7th (the correct final date will be communicated as soon as possible), before the main conference begins.
ICDAR 2021 Tutorials should serve one or more of the following objectives:
- Introduce students and newcomers to major topics of Document Analysis and Recognition (DAR) research.
- Provide instructions on established practices and methodologies.
- Introduce expert non-specialists to a DAR subarea. Survey a mature area of DAR research and/or practice.
- Motivate and explain a DAR topic of emerging importance.
- Overview DAR systems for industrial solutions (suggestion for researchers in industry).
- Introduce some recent innovative techniques for DAR research and software quality, such as open-source libraries, high-level API, technical frameworks for expert developments, etc. (suggestion for expert programmers).
An ICDAR tutorial should aim to give a comprehensive overview of a specific topic related to DAR. A good tutorial should be educational rather than just a cursory survey of techniques. The topic should be of sufficient relevance and importance to attract significant interest from the ICDAR community. Typical tutorial audiences consist of PhD students studying computer vision, image processing or pattern recognition, but also include researchers and practitioners from both academia and industry.
In order to facilitate innovative collaboration and interaction between researchers in academia and industry, the Tutorial Chairs strongly encourage proposals for industrial tutorials, in which researchers in companies describe DAR systems and overview industrial solutions to document analysis problems in real use-case industrial scenarios.
Proposals should be up to 4 pages in length, and should contain the following information:
- Title of the tutorial.
- Scope and motivation. A brief description of the tutorial, suitable for inclusion in the conference registration brochure.
- Preference for the duration (full day or half day). Due to agenda constraints, half day tutorials are recommended. If a full day is needed, provide a brief justification.
- A detailed outline of the tutorial. Course description with list of topics to be covered, along with a brief outline.
- Relevance for ICDAR. A description of why the tutorial topic would be of interest to a substantial part of the ICDAR audience.
- Expected target audience in terms of composition and estimated number of attendees. Prerequisite knowledge of the ICDAR audience for attending the tutorial.
- Short CV of organizers. A brief CV of the presenter(s), including name, postal address, phone number, e-mail address, web page, background in the tutorial area (projects, relevant publications or tutorial-level articles on the subject), evidence of teaching experience.
- The name and e-mail address of the corresponding presenter. The corresponding presenter should be available for e-mail correspondence during the evaluation process, in the case clarifications and discussions on the scope and content of the proposal are needed.
The evaluation of the proposal will take into account its general interest for ICDAR attendees, the quality of the proposal (e.g., a tutorial that simply lists a set of concepts without any apparent rationale behind them will not be approved) as well as the expertise and skills of the presenters. We emphasize that the primary criteria for evaluation will be whether a proposal is interesting, well-structured, and motivated in relation to Document Analysis and Recognition, rather than the perceived experience/standing of the proposer.
Last but not least, the tutorial should attract a meaningful audience, cover hot topics and incorporate new knowledge to the community. Those submitting a proposal should keep in mind that tutorials are intended to provide an overview of the field; they should present reasonably well established information in a balanced way. Tutorials should not be used to advocate a single avenue of research, nor should they promote a product.
- Tutorial slides must be provided to us for inclusion on the conference website and also on the TC-10 and TC-11 websites, as educational material.
- The ICDAR main conference organizers will handle the tutorial registration and provide the space, coffee breaks and other facilities required to organize tutorials (e.g. a room, a projector and a screen).
Submission Guidelines & Inquiries
All proposals should be submitted by electronic mail to the Tutorial Chairs:
Feedback, comments and/or suggestions would be provided within two weeks of receiving the proposal. Final acceptance (or rejection) would be decided by December 23, 2020.
For any inquiries you may have, please contact us via the above emails.
ICDAR 2021: Workshops
During the pre-conference from September 5-7, 2021, several workshops related to document analysis and recognition will be held in conjunction with the ICDAR 2021 main conference.
- ICDAR 2021 Workshop on Graphics Recognition (GREC 2021, 14th edition)
- ICDAR 2021 Workshop on Camera-Based Document Analysis and Recognition (CBDAR 2021, 9th edition)
- ICDAR 2021 Workshop on Historical Document Imaging and Processing (HIP 2021, 6th edition)
- ICDAR 2021 Workshop on Arabic and Derived Script Analysis and Recognition (ASAR 2021, 4th edition)
- ICDAR 2021 Workshop on Computational Document Forensics (IWCDF 2021, 3rd edition)
- ICDAR 2021 Workshop on Future of Document Analysis and Recognition (FDAR 2021, 3rd edition)
- ICDAR 2021 Workshop on Human-Document Interaction (HDI 2021, 3rd edition)
- ICDAR 2021 Workshop on Machine Learning (WML 2021, 3rd edition)
- ICDAR 2021 Workshop on Open Services and Tools for Document Analysis (ICDAR-OST 2021, 3rd edition)
- ICDAR 2021 Workshop on Industrial Applications of Document Analysis and Recognition (WIADAR 2021, 2nd edition)
- ICDAR 2021 Workshop on Computational Paleography (1st edition)
- ICDAR 2021 Workshop on Document Image and Language (WDIL 2021, 1st edition)
- ICDAR 2021 Workshop on Document Visual Question Answering (DocVQA 2021, 1st edition)
- ICDAR 2021 Workshop on Graph Representation Learning for Scanned Document Analysis (GLESDO 2021, 1st edition)
For any question, please contact the Workshop Chairs, Elisa Barney Smith and Umapada Pal:
ICDAR 2021: Long-Term Competition on Document Visual Question Answering (DocVQA)
- NEW DATASETS ARE NOW AVAILABLE ***
We are pleased to invite you to participate in the ICDAR 2021 Long-Term Competition on Document Visual Question Answering (DocVQA) that is organized in conjunction with the 16th International Conference on Document Analysis and Recognition (ICDAR), 5-10 September 2021, Lausanne (Switzerland).
Contemporary Document Analysis and Recognition (DAR) research tends to focus on generic information extraction tasks (character recognition, table extraction, word spotting), largely disconnected from the final purpose the extracted information is used for. The DocVQA challenge seeks to inspire a “purpose-driven” point of view in Document Analysis and Recognition research, where the document content is extracted and used to respond to high-level tasks defined in natural language by the human consumers of this information. In this sense DocVQA constitutes a new problem space, where high-level semantic tasks dynamically drive information extraction algorithms to conditionally interpret document images.
DocVQA is modelled as a Visual Question Answering (VQA) problem where the task is to answer a natural language question asked on a single document image or a collection of digitised documents. This goes over and above passing a document image through OCR, and involves understanding all types of information conveyed by a document.
The DocVQA challenge is a continuous, evolving effort, linked to various scientific events. The ICDAR 2021 competition is the second edition of the challenge, while a first edition was organized in the context of CVPR 2020 Workshop on Text and Documents in the Deep Learning Era.
The challenge is structured around three tasks, out of which Task 2 and Task 3 constitute the core of the ICDAR 2021 edition:
Task 1: Single Document VQA. VQA over a generic set of ~40k questions over ~10k documents. Introduced for CVPR 2020, and continuously open for submissions.
Task 2: Document Collection VQA. VQA over a collection of ~14k forms. Locating the evidence is as important as finding the correct answer. Introduced for CVPR 2020 and revamped for ICDAR 2021.
Task 3: Infographics VQA. VQA over challenging documents with complex layout. New task introduced for ICDAR 2021 on a new dataset of ~5k documents
The competition is hosted at the Robust Reading Competition portal. Please register to the portal in order to gain access to downloads and submit results.
10 November 2020: Release of training data subset for new Task 3 on “Infographics VQA”
10 December 2020: Release of full dataset for new Task 3 on “Infographics VQA”
31 March 2021: Deadline for Competition submissions
5-10 September 2021: Results presentation at ICDAR
Minesh Mathew (IIIT Hyderabad, India)
Rubèn P. Tito (Computer Vision Center, Spain)
C. V. Jawahar (IIIT Hyderabad, India)
Dimosthenis Karatzas (Computer Vision Center, Spain)
ICDAR 2021: Competition on Historical Map Segmentation
We would like to encourage you to participate in the ICDAR 2021 Competition on Historical Map Segmentation, to be organized in conjunction with the 16th International Conference on Document Analysis and Recognition (ICDAR), 5-10 September 2021, Lausanne (Switzerland).
This competition is organized by the LASTIG team of IGN (the French National Mapping Agency), the R&D Lab. of EPITA (French engineering school in computer science), and the Center of Historical Studies of EHESS (French graduate schools of social sciences).
Why should you consider this competition? All datasets are released with an open licence as soon as they are available (training and validation sets already available). Evaluation tools are released early with an open source licence. Winners of each task will be invited to co-author the report paper. We propose unsolved research problems at the frontiers of computer sciences, social sciences and history, with an important potential impact in the geographic information systems (GIS) community.
Downloads, mailing list, updates, evaluations tools available here: https://icdar21-mapseg.github.io/
Twitter account: https://twitter.com/ICDAR21_MapSeg
Contact address: icdar21-mapseg-contact(at)googlegroups.com
Task 1: detecting a set of closed shapes (building blocks) in the map image.
Task 2: segmenting the map content from the rest of the sheet.
Task 3: locating the intersection points of graticule lines (historical georeferencing lines).
Now: training and validation sets available - Training phase is started
March 31, 2021: Registration deadline for competition participants
April 5 to April 16, 2021: Test phase
April 23, 2021: Notification of results, Method description deadline
July 1, 2021: Full dataset release, Full results disclosure.
Edwin Carlinet (LRDE/EPITA)
Joseph Chazalon (LRDE/EPITA)
Yizi Chen (LASTIG/IGN & LRDE/EPITA)
Bertrand Duménieu (CRH/EHESS)
Thierry Géraud (LRDE/EPITA)
Clément Mallet (LASTIG/IGN)
Julien Perret (LASTIG/IGN & CRH/EHESS)
TC11 Datasets Repository (repost)
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)
( email@example.com )
PostDoc-position in Computer Vision and Machine Learning for Document Image Fraud Detection
L3i laboratory at La Rochelle, France
Start date: February 1st, 2021 (may be modified according to the sanitary situation)
Duration: 12 months
Income: 2100 € / month
The work carried out by the candidate will be a part of the LabCom IDEAS. IDEAS is co-funded by the French National Research Agency (ANR) and the Nouvelle-Aquitaine Region, and brings together the Yooz company and the L3i laboratory. The objective is to imagine, invent, design, develop, optimise and train the best algorithms for processing automatically business documents. The goal is to offer services, based on artificial intelligence, that are capable of automatically analysing and understanding various types of business documents.
The work of the postdoc fellow will fall within the area “Document fraud detection”. The aim is to develop new methods for fraud detection in document images (image forensics) without a priori knowledge. The candidate, who holds a Ph.D. in the fields of computer science, computer engineering, signal processing, or applied mathematics, must have a significant research experience in the domain of image or signal processing and analysis.
Candidates for this position should send a CV and a cover letter (names and reference details would be appreciated) to:
- petra.gomez [at] univ-lr.fr
- mickael.coustaty [at] univ-lr.fr
- nicolas.sidere [at] univ-lr.fr
- Vincent.PoulaindAndecy [at] getyooz.com
- Saddok.Kebairi [at] getyooz.com
More details available at: https://www.madics.fr/event/offre206/
Call for Contributions: To contribute news items, please send a short email to the editor, Andreas Fischer (). 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.