Everyone without registration is welcome.
The Workshop on the Future of Document Analysis and Recognition
in cooperation with the Institute of Document Analysis and Knowledge Science (IDAKS), Osaka Prefecture University
Date: November 12, 2017 (A half day, PM)
Location: Large Conference Room, 3rd Floor
Call for Participation
Workshop Chairs
- Koichi Kise, Osaka Prefecture University, Japan
- Andreas Dengel, DFKI, Germany
Objective
Let’s discuss the future of document analysis and recognition. In the history of ICDAR in the last 26 years, we have been working in a variety of fields of document analysis and recognition, and have solved many important problems. The emergence of machine learning technologies, however, has shifted the nature of unsolved problems. For example, improving accuracy of methods for existing problems attracts less researchers. This brings us to the following questions: What are exciting open problems that interest more researchers? What are research topics young researchers and outsiders of the ICDAR community cannot help joining us?
This workshop is planned to discuss possible research topics and open problems to start new chapters of research on document analysis and recognition.
Workshop format
The purpose of the workshop is to have deeper discussions about the future of document analysis and recognition. For stimulating the discussion, we have short presentation by invited speakers listed below to share their views of the future. After the presentation, we select some representative topics for discussion. Then participants and invited speakers are divided into some groups of selected topics for discussion. The results of discussion will be summarized at the end of the workshop and final results will be presented at the main conference. We plan to have a short summary presentation at the banquet and more detailed presentation and discussion at the panel on the last day of the conference.
Invited Speakers
- Apostolos Antonacopoulos
- Michael Blumenstein
- Andreas Dengel
- Dimosthenis Karatzas
- Koichi Kise
- C. V. Jawahar
- Bart Lamiroy
- Cheng-Lin Liu
- Marcus Liwicki
- Josep Llados
- Dan Lopresti
- Simone Marinai
- Jean-Marc Ogier
- Seiichi Uchida
Program
14:30-14:40 Opening
14:40-15:50 Position paper presentation (5 min/invited speaker)
15:50-16:20 coffee break
16:20-16:30 Discussion group formation (about 4 groups)
16:30-17:50 Discussion (each group)
17:50-18:30 wrap-up (5 min/group + summary)
Abstracts of Position Papers
-
Apostolos Antonacopoulos:
Context-Aware Document Analysis -
Michael Blumenstein:
Beyond Deep Learning – the next big thing in Artificial Intelligence for Document Analysis Research -
Andreas Dengel:
Who are we now? A Multi-Perspective View to Documents as well as Document Analysis and Recognition -
Dimosthenis Karatzas:
Large scale datasets – Is data holding us back? -
Koichi Kise:
Mutual Analysis of Documents and their Readers/Writers -
C. V. Jawahar:
Towards Deeper Understanding of Wider Category of Documents -
Bart Lamiroy:
Document Analysis and Machine Learning ... beyond simply fancy classification? -
Cheng-Lin Liu:
Document Image Analysis: Big Data or Small Data -
Marcus Liwicki:
Really Open Services and Tools for Reproducible Research -
Josep Llados:
From OCR (Optical Character Recognition) to IDS (Intelligent Decoding Systems): an edit distance of 3, a time span of 100. -
Dan Lopresti:
Document Analysis and Recognition: Taking the Pulse of Our Field -
Simone Marinai:
Document Image Analysis and Pattern Recognition, what is their position today? -
Jean-Marc Ogier:
Which challenges for DAR in the context of a hybrid world ? -
Seiichi Uchida:
Beyond 100%
Looking forward to meeting you at the workshop!!