Offline continuous handwriting recognition using sequence to sequence neural networks

Sueiras, Jorge; Ruíz, Victoria; Sánchez, Ángel Mediavilla; Vélez, José Francisco

doi:10.1016/j.neucom.2018.02.008

Cited by 133 publications

(73 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This state is used to obtain a value for each element of the sequence that will be added to the values of the sequence. This diagram reproduces the attention mechanism of [2]. Adjusting the Perceptron, responsible for the weight of the mechanism, we achieved very interesting results, as displayed in Figure 5 where we see the attention which follows the figures.…”

Section: E Retained Architecturesupporting

confidence: 64%

“…[1] and motivated by its successful application on handwritten word recognition by Sueiras and al. [2]. As it will be explained later, our system is able to transform a variable-length sequence of pixel columns, extracted from the handwritten digit string image, into a variable-length sequence of digits to form a numerical string.…”

Section: Proposed Approachmentioning

confidence: 99%

“…The problem is that we do not usually know the number of digits in the string and so the optimal boundary between them is unknown. Such a problem has been dealt with in different ways [5], [9], [14] and one way to approach it is to see it as a sequence-to-sequence problem with a sequence of image patches as input and a sequence of characters as output [2]. To solve sequenceto-sequence problems with different length from input to output, the key point is to find the alignment between them.…”

Section: Introductionmentioning

confidence: 99%

“…In the prior probabilities based alignment, the attention mechanism is most often used in sequence-to-sequence models to learn many-to-one alignment [8]. Moreover, it had led to better results than HMM and CTC, for text translation [8], speech recognition [11] and word recognition [2]. Combining the attention mechanism with HMM or CTC might also lead to improvements [15].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

On the Use of Attention Mechanism in a Seq2Seq Based Approach for Off-Line Handwritten Digit String Recognition

Lupinski

Belaïd

Echi

2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

In this work, we investigate the use of the attention mechanism in deep learning for a better reading of handwritten digit strings in digitized images. The proposed recognition system built upon a CNN (Convolutional Neural Network) and two RNNs (Recurrent Neural Networks), acting as Encoder and Decoder and using the attention mechanism. We used a 1D mechanism for attention location with a "soft" alignment attention which has the peculiarity of having an easily calculable gradient and thus to integrate well with the network. Experimental results on data from ORAND-CAR A, ORAND-CAR B and CVL HDS databases compare favorably to other published methods.

show abstract

Section: E Retained Architecturesupporting

confidence: 64%

Section: Proposed Approachmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On the Use of Attention Mechanism in a Seq2Seq Based Approach for Off-Line Handwritten Digit String Recognition

Lupinski

Belaïd

Echi

2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

View full text Add to dashboard Cite

show abstract

“…In particular, [12] and [13] use BLSTMs for the recurrent encoder. Some works, like [14], [15], use similar architectures, but limit their work on recognizing isolated handwritten words. A bidirectional decoder is incorporated in [24], by integrating a length estimation procedure.…”

Section: Related Workmentioning

confidence: 99%

Evaluating Sequence-to-Sequence Models for Handwritten Text Recognition

Michael

Labahn

Grüning³

et al. 2019

2019 International Conference on Document Analysis and Recognition (ICDAR)

104

View full text Add to dashboard Cite

Encoder-decoder models have become an effective approach for sequence learning tasks like machine translation, image captioning and speech recognition, but have yet to show competitive results for handwritten text recognition. To this end, we propose an attention-based sequence-to-sequence model. It combines a convolutional neural network as a generic feature extractor with a recurrent neural network to encode both the visual information, as well as the temporal context between characters in the input image, and uses a separate recurrent neural network to decode the actual character sequence. We make experimental comparisons between various attention mechanisms and positional encodings, in order to find an appropriate alignment between the input and output sequence. The model can be trained end-to-end and the optional integration of a hybrid loss allows the encoder to retain an interpretable and usable output, if desired. We achieve competitive results on the IAM and ICFHR2016 READ data sets compared to the state-of-theart without the use of a language model, and we significantly improve over any recent sequence-to-sequence approaches.

show abstract

Convolve, Attend and Spell: An Attention-based Sequence-to-Sequence Model for Handwritten Word Recognition

Kang

Toledo

Riba

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Offline continuous handwriting recognition using sequence to sequence neural networks

Cited by 133 publications

References 11 publications

On the Use of Attention Mechanism in a Seq2Seq Based Approach for Off-Line Handwritten Digit String Recognition

On the Use of Attention Mechanism in a Seq2Seq Based Approach for Off-Line Handwritten Digit String Recognition

Evaluating Sequence-to-Sequence Models for Handwritten Text Recognition

Convolve, Attend and Spell: An Attention-based Sequence-to-Sequence Model for Handwritten Word Recognition

Contact Info

Product

Resources

About