Offline recognition of unconstrained handwritten texts using HMMs and statistical language models

By Alessandro Vinciarelli, Samy Bengio, Horst Bunke. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004

341 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Improving Offline Handwritten Text Recognition with Hybrid HMM/ANN Models (2011, background): “[6], [9], [10], [11] after their success in automatic speech”
Bidirectional Language Model for Handwriting Recognition (2012, background): “A straightforward way to build both lists is to use a recognizer for handwritten text that produces a recognition lattice, such as HMMs or BLSTM Neural Networks in conjunction with a Token Passing algorithm.”
Segmentation (2019, background): “…or blood vessels; homogeneous densities correspond to ﬁbrous tissue which appears as bright areas in mammographic images, hides the underlying normal TDLU, ducts and their alterations; radiolucent areas are related to adipose fatty tissue which appears as dark areas in mammographic images [3].”
ICDAR 2013 Chinese Handwriting Recognition Competition (2013, methodology): “A correct rate (CR) and an accurate rate (AR) [5,6] are calculated over all the text lines in the test dataset:”
The A2iA Arabic Handwritten Text Recognition System at the Open HaRT2013 Evaluation (2014, background): “The importance of using statistical n-gram language models in handwritten recognition systems has been demonstrated in several studies [14] [15].”
Offline Arabic handwriting recognition: a survey (2006, neutral): “not in the lexicon [94, 95].”