Get to Know Dengel

After finishing my studies of Computer Science and Economics at the University of Kaiserslautern in 1986, I started working at Siemens Labs in Munich, Germany, where I did research on Geometric Reasoning and Pattern Recognition. Following my first successful publications at IJCAI and ICPR, I received an offer as a lecturer and research assistant at the Computer Science faculty of the University of Stuttgart where I finished my PhD in February, 1989.

During my early years, in March, 1989, I met Horst Bunke in Boston and Cambridge, MA, where I gave a talk at MIT. Horst told me about the SSPR 90 workshop held in Murray Hill, NJ, organized by Henry Baird and Larry O’Gorman and the idea of bringing together everyone who worked in the field of Document Analysis and Recognition. For me it was an absolute must, and finally I was able to successfully submit a paper. Today, I can say that SSPR 90 was a kind of key event for me: I was not only inspired by the atmosphere and the constructive discussions but was also able to get networked to many distinguished community members. In this way, I got in touch with all those senior people who were involved in the early planning and the setting-up of the ICDAR conference series. To my surprise, I was invited to become a PC member of the first ICDAR, which took place in Saint-Malo, France, and of all other ICDARs that followed.

When coming back to Germany from a temporary, post-doctoral position at PARC, I started working at the German Research Center for Artificial Intelligence (DFKI) and set-up my own small group that focused on various topics of document analysis and recognition. In 1993, I was invited by Kazuhiko Yamamoto and Yasuaki Nakano to give a keynote talk at ICDAR 1993, my first of three keynotes at ICDAR conferences. In the same year I was appointed professor at the University of Kaiserslautern and took over the job as a Scientific Director at DFKI. This provided the opportunity to work on Document Analysis and Recognition within an excellent and well-staffed institution.

Taking the basic idea of SSPR90, Henry Baird, Larry Spitz, and myself decided to run a new but special IAPR biannual workshop on Document Analysis Systems (DAS), which was to take into consideration the increasing demand from the market on such technologies. As a consequence, the first DAS took place in my hometown, Kaiserslautern, Germany, in 1994. At the same time, my group at DFKI concentrated more and more on the combination of Knowledge-Based and Pattern Recognition Methods for understanding the contents of various kinds of office documents. The aim of our research was to extract all workflow-relevant bits of information and route them to post-processing applications. This research addressed a new field and required new approaches. Building on George Nagy’s early publications on X-Y Trees (see Getting to Know...George Nagy, IAPR Fellow in this issue), my group and I could develop very promising techniques for document understanding and information extraction. At the same time and together with Jürgen Schürmann, one of the pioneers in Character Recognition, we succeeded in hosting ICDAR 97 in Ulm, Germany, and I acted as a Program Chair. During this conference IAPR introduced the ICDAR Young Investigator Award, which I was honored to receive together with J.J. Hull. In Ulm, we started to discuss ideas to run our own journal, the IJDAR, published by Springer, that would further help to foster collaboration and the exchange of ideas within our community. I was an editor right from the beginning, which was a great honor for me. The first issues started in 1998.

During these years, my group at DFKI rapidly grew, with a broad range of international publications as well as several very successful industrial projects. This led to my first spin-off company built on our Document Analysis and Recognition technology workbench. This was in 1999. Today, Insiders Technologies is well established with more than a hundred employees and installations all over the world.

In the meantime, ICDAR became a major event for the whole community. Successful conferences in Tsukuba City, Montreal, and Bangalore followed Saint-Malo, until we all met again in September 2001 in Seattle. It was a special ICDAR, but not only because of 9-11 that happened during the conference and overshadowed everything, but this event revealed that such a large conference needed a forum for continuous monitoring and for providing appropriate advice for planning, organizing, and running coming ICDAR conferences. This was the birth of the ICDAR Advisory Board to which TC10 and TC11 chairs and one or two senior people would belong. I was elected to join, and today, I am still a (the longest-serving) member.

In 2004, I ran another DAS as a Co-Chair together with Simone Marinai in Florence, and for the first time, we had more than one hundred participants. At ICPR in Cambridge, UK, I was elected an IAPR Fellow for my contributions to document understanding and for service to IAPR. One year later, I founded another start-up, again on Document Analysis, and received the Pioneer Spirit Award, a prize for the high quality of the underlying technological and founding concept. In 2006, we had another DAS in Nelson, New Zealand, chaired by Larry Spitz and Horst Bunke. When Larry Spitz decided to retire that year, I took over his role as a spokesman of the DAS Steering Committee. In the fall of the same year, I acted as a Track Chair at ICPR in Hong Kong. In 2007, Yasuaki Nakano, one of our distinguished Japanese colleagues and honorary chair of DAS 2008, became seriously ill. Unfortunately he didn’t recover and died of cancer. In memory of him, IAPR announced the Nakano Award for the best paper at DAS workshops. I was deeply moved when Nakano-sensei’s widow handed the first Nakano Award to my student Georg Buscher and me at DAS 2008 in Nara.

Over the years, I have not only established research alliances but also strong personal relations to Japan. Among the many ties, I have to mention Koichi Kise, who was a visiting researcher at DFKI for 15 months at the turn of the millennium. It was also him with whom I founded the Institute on Document Analysis and Knowledge Science (IDAKS) at the Osaka Prefecture University (OPU) in 2008. Based on my frequent visits to OPU and the close collaborations with jointly supervised theses and continuous exchange, in 2009, I was highly honored to become the first non-Japanese to be appointed a Kyakuin (Honorary Professor) at OPU, which made me very proud.

Today, I am a member of the management board of DFKI, the world’s largest AI institute with more than 800 employees, an exciting place with exciting people, and one of the best places I can think of to do research. In all these years, my work was always been dedicated to knowledge-based and cognitive aspects on pattern recognition with a strong focus on applications, especially on documents. Beside the traditional fields, I am currently experimenting a lot with perception and eye tracking in order to determine attention. For those who are interested, please see Text 2.0 (http://text20.net/).

There are, God willing, still some fifteen years left in my professional life. This year, I am involved in organizing ICPR 2012 in Tsukuba, Japan as a Tutorial Chair and in ICFHR in Bari as a Program Co-Chair. I am quite sure that there will be a whole bunch of new, exciting contributions that I can make to our Pattern Recognition society in the future.