DAR Strategy

Document Analysis and Recognition (DAR), the research discipline that aims to replicate in machines the ability to read, has been one of the pioneering fields in artificial intelligence and computer vision for the past 50 years or so. At the meeting point between vision and language, DAR covers important research and application areas ranging from administrative document analysis and digital libraries to scene text understanding and music recognition. As such it has acted as a connection point between academia and industry, as well as for researchers of different disciplines, including computer vision, natural language processing, digital humanities, and human computer interaction to mention just a few.

The community is mainly gathered around two Technical Committees of the Int. Association for Pattern Recognition, TC10 on Graphics Recognition (dealing with the understanding of graphical languages and graphical elements) and TC11 on Reading Systems (dealing with the understanding of written communication in images). The two TCs have a combined membership of above 1,300 DAR researchers and practitioners, and run a series of well known events in the field: the Int. Conf. on Document Analysis and Recognition (ICDAR), the Int. Conf. on the Frontiers of Handwriting Recognition (ICFHR), the Document Analysis Systems workshop (DAS), the TC10/TC11 Summer School and the TC10/11 Doctoral Consortium.

The Document Analysis and Recognition community has been evolving continuously, and is now as relevant as it was when the artificial intelligence term was coined back in the 50s. The community has traditionally been very open and participatory. Since ICDAR 2017, we have organized a series of activities that have allowed us to reflect on numerous aspects, from research quality to organization matters.

At the TC10/11 meeting at ICDAR 2019 it was decided to set up a small and agile Task Force to work on a mid-long term strategy for the community. The Task Force was created at the beginning of 2020, and actually managed to do a first face-to-face meeting back in January – just before the world closed down. Since then it has been working online and the TC10 and Tc11 chairs gave short progress updates during DAS 2020 and ICFHR 2020.

This Web site was setup to collect all the information related to this process, and facilitate wider participation.

Alicia Fornes (TC10 Chair), Dimosthenis Karatzas (TC11 Chair)

December 2020

What can you do now? Get a better idea about how all this came around and how it has evolved over the past few years by visiting the Timeline. Have a look at all the shared Resources, check out in particular at the data analysis to get a feeling of where we come from and the current version of the strategic plan. Participate in the ongoing Discussions. And above all, give us Feedback on the strategic plan! Is there anything we have missed? Do you feel strongly about a particular action? Do you like all you see? Please share your thoughts with us.