To spot keywords on handwritten documents, we present a hybrid keyword spotting system, based on features extracted with Convolutional Deep Belief Networks and using Dynamic Time Warping for word scoring. Features are learned from word images, in an unsupervised manner, using a sliding window to extract horizontal patches. For two single writer historical data sets, it is shown that the proposed learned feature extractor outperforms two standard sets of features.
CITATION STYLE
Wicht, B., Fischer, A., & Hennebert, J. (2016). Keyword spotting with convolutional deep belief networks and dynamic time warping. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9887 LNCS, pp. 113–120). Springer Verlag. https://doi.org/10.1007/978-3-319-44781-0_14
Mendeley helps you to discover research relevant for your work.