ICPR2016 Conference – Deep Learning Features for Handwritten Keyword Spotting

iCoSys, in collaboration with DIVA , presented on 8th of December 2016 a scientific communication entitled “Deep Learning Features for Handwritten Keyword Spotting” at the International Conference on Pattern Recognition – ICPR2016 in Cancun, Mexico.

Congrats to the authors for the acceptation of this paper.

Abstract: Deep learning had a significant impact on diverse pattern recognition tasks in the recent past. In this paper, we investigate its potential for keyword spotting in handwritten documents by designing a novel feature extraction system based on Convolutional Deep Belief Networks. Sliding window features are learned from word images in an unsupervised manner. The proposed features are evaluated both for template-based word spotting with Dynamic Time Warping and for learning-based word spotting with Hidden Markov Models. In an experimental evaluation on three benchmark data sets with historical and modern handwriting, it is shown that the proposed learned features outperform three standard sets of handcrafted features.

Download the presentation.


Leave a Reply