Research

By JP van Oosten

As a PhD student, I worked on the Monk system, a search engine for historical, handwritten manuscripts. I studied hidden Markov models and other techniques for handwriting recognition.

Thesis

The snowball principle for handwritten word-image retrieval
The importance of labelled data and humans in the loop

I defended my thesis on March 26th, 2021. Feel free to download it here.

Awards

• IAPR Best Oral Paper Award at ICFHR 2014, (van Oosten & Schomaker, on 'A reevaluation and benchmark of hidden Markov models')
• IAPR Best Paper Award at ICFHR 2012, (van Oosten & Schomaker, on 'Separability versus prototypicality in handwritten word retrieval')

Publications

• J.P. van Oosten, and L.R.B. Schomaker. "A reevaluation and benchmark of hidden Markov models." Proc. Int. Conference on Frontiers in Handwriting Recognition, September 2014, Crete, Greece, IEEE Computer Society, pp. 531-536. Available: 10.1109/ICFHR.2014.95
• Van Oosten, Jean-Paul, and Lambert Schomaker. "Separability versus prototypicality in handwritten word-image retrieval." Pattern Recognition 47.3 (2014): 1031-1038. Available: 10.1016/j.patcog.2013.09.006
• Niitsuma, Masahiro, et al. "Writer Identification in Old Music Manuscripts Using Contour-Hinge Feature and Dimensionality Reduction with an Autoencoder." Computer Analysis of Images and Patterns. Springer Berlin Heidelberg, 2013. Available: 10.1007/978-3-642-40246-3_69
• J.P. van Oosten, and L.R.B. Schomaker (2012). "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.
• T. K. Bhowmik, J.P. van Oosten, and L. Schomaker, "Segmental K-Means learning with mixture distribution for HMM based handwriting recognition pattern recognition and machine intelligence," ser. Lecture Notes in Computer Science, S. O. Kuznetsov, D. P. Mandal, M. K. Kundu, and S. K. Pal, Eds. Berlin, Heidelberg: Springer Berlin / Heidelberg, 2011, vol. 6744, ch. 70, pp. 432-439. [Online]. Available: 10.1007/978-3-642-21786-9_70
• J.P. van Oosten, "Can Markov properties be learned by hidden Markov modelling algorithms?", 2010, Master's thesis, University of Groningen, the Netherlands