prof. dr. Lambert Schomaker
Artificial Intelligence and Cognitive Engineering (ALICE)
[Research | Education]
Bernoulliborg building (V)
9747 AG Groningen, The Netherlands
E-mail: home email address
Researchers at ALICE
APS group meetings
Here is a link to the group meetings of the Autonomous Perceptive Systems group.
Interests & projects
- Within artificial intelligence, my focus is on perceptual intelligence and (lifelong) machine learning
- Monk is our continuous project for allowing access to large and diverse
historical archives which are difficult to access by means
of traditional 'OCR' approaches due to their special fonts or handwritten style.
Since 2009, Monk provides a means of continuously training handwriting recognizers that create indices of handwritten
collections. Monk was one of the demonstrators in the EU/SNN Target project
- May 2019: EU/ITN Manic project approved, participation: ML and neuromorphic computing / cognitive materials
- 2018 EU/ITN Perico project participation: Machine learning for peroxisome classification
- Dead Sea Scrolls ERC project with Mladen Popovic (PI) and Hans van der Plicht (14C dating)
- August 2015: check out our new Pattern Recognition article on Chinese/English cross-script writer identification using junclets.
- March 2015, we obtained, together with Dominique Stutzmann the NWO project HIMANIS, HIstorical MANuscript Indexing for user-controlled Search in order to test our methods for word retrieval on the French royal chancery (14th-15th c.), tackling new problems in image processing, word segmentation and allographic variation
- January 2015, we obtained an ECSEL/H2020 grant, as a member of a consortium of 47 industrial and research partners, concerning big data, machine learning
and preventive maintenance in (robotic) manufacturing lines: project Mantis
- January 2014, an international NWO/Digging into Data project was granted titled
"Global Currents" in which Stanford University and McGill University participate.
The application domain is e-Humanities and our contribution to the project is the induction of semantics from visual patterns, shapes and
their occurrence on historical (handwritten) documents, using pattern recognition and machine learning.
- In 2012, we started a project in the computer-based dating of mediaeval handwritten manuscripts, in a Dutch NWO project "MPS - The Mediaeval Paleographic Scale", together
with paleographer prof. dr. Jan Burgers from the Huygensinstituut, The Hague. We will use models and methods collected in the area of writer identification.
- Ubbo Emmius' RERUM FRISICARUM HISTORIA (1616) Machine-print OCR on an important historical book (a GUF grant with Zweder von Martels & Arnold Meijster)
- NWO/Morph: Learning to learn: An Adaptive Reading System using a High-Performance Morphed-Image Correlator
- NWO/CATCH, sub project Scratch (SCRipt Analysis Tools for the Cultural Heritage):
Historical document analysis, together with the Nationaal Archief.
- Biologically inspired image features (Poggio/Serre) are strong in whole-word recognition
- Off-line writer identification
- Our recognizer of mixed-styles online handwriting
- Methods: (Deep) Neural networks, CNN/LSTM, SVMs, Multiple agents, Genetic algorithms, finding proper distance measures for nearest-neighbour search
- Content and
object-based image search
- Service robotics (also see: Mantis project)
- Autonomous Systems / camera-based text detection for robots and smart phones
- People in my group
- Former M.Sc. Students
- Ph.D. involvement
- Other committees
Researchers, postdocs and scientific programmers
- Sukalpa Chanda (postdoc Himanis project) writer identification and curvilinear line segmentation, deep learning for word recognition
- Tapan Bhowmik (postdoc Target project, HMMs & historical manuscript recognition)
- Arnold Meijster (high-performance computing, image processing)
- Aswin van Woudenberg (writer identification/implementation of GIWIS)
- Edzer Lawerman (database and I/O optimization, seconded, NSpyre)
- Sha Luo Reinforcement-based learning of object handling in robots
- Mahya Ameryan Deep learning for text recognition and pictorial semantics
- Maruf Dhali Dead Sea Scrolls, document dating and writer identification
- Jean-Paul van Oosten (Markov modeling of active recognition)
He has obtained, at the ICFHR 2012, the IAPR Best Paper Award:
Jean-Paul Van Oosten, Lambert Schomaker (2010).
Separability versus Prototypicality in Handwritten Word Retrieval,
Proc. Int. Conference on Frontiers in Handwriting Recognition, September 18-20 2012, Bari, Italy, IEEE Computer Society, pp. 8-13,
- Bowornrat Sriman (camera-based text detection and recognition for Asian scripts)
- Amir Shantia Robotics perception and navigation (together with Marco Wiering)
- Sheng He (writing style-based dating of handwritten manuscripts, dissertation 2017, 'cum laude' - with honors)
- Marius Bulacu (writer identification, dissertation 2007)
- Tijn van der Zant (handwritten historical document retrieval, dissertation 2010)
- Axel Brink (robust writer identification and verification, dissertation 2011)
- Gert Kootstra (biologically inspired robot vision and navigation, dissertation 2010)
- Hado van Hasselt (reinforcement learning systems, with Marco Wiering, dissertation 2011)
- Maikel Withagen (LSTMs and natural-gas network turbine load prediction)
- Marianne Ritsema van Eck (semantic modeling of 15th century Alderman's acts for improval of handwriting recognition)
- M.Sc. project topics
Setare Rezaee is a PhD guest, working on Persian (Farsi) handwriting recognition using deep learning (2018/2019).
dr. Gideon Maillette de Buy Wenniger is our current guest postdoc (2018).
Masahiro Niitsuma has visited our lab for his research project on
musical score identification.
dr. Nobuo Ezaki has visited our lab. He obtained
a Japanese grant for research in developing an experimental camera-based text
reading system in order to be able to realize a text-to-speech
conversion for blind people who are navigating a typical city or office
environment with text tags.
Tusar Mishra Kanti has visited our lab, working in Odia and Bangla
character recognition. These scripts have highly variable
'pig-tailed' initial and ending strokes that make the notion of a stable
rectangular character bounding box questionable. This necessitates the use of
special shape features or character-recognition strategies.
I heartily welcome International Students to
take a look at our programmes.
Here is brochure on the institute ALICE of which I am
the scientific director. My corporate web site is here