Offline Handwriting Recognition with Multidimensional Recurrent Neural Networks

By Alex Graves, J. Urgen Schmidhuber. 2008

444 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Human-Computer Interaction

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Human-Computer Interaction

Citations in the corpus

Generating Sequences With Recurrent Neural Networks (2013, background): “Long Short-term Memory (LSTM) [16] is an RNN architecture designed to be better at storing and accessing information than standard RNNs. LSTM has recently given state-of-the-art results in a variety of sequence processing tasks, including speech and handwriting recognition [10, 12].”
Conditional Image Generation with PixelCNN Decoders (2016, methodology): “Two variants were proposed in the original paper: PixelRNN, where the pixel distributions are modeled with two-dimensional LSTM [7, 26], and PixelCNN, where they are modelled with convolutional networks.”
Generative Image Modeling Using Spatial LSTMs (2015, methodology): “In the following, we first review and extend the MCGSM [38] and multi-dimensional LSTMs [8] before explaining how to combine them into a recurrent image model.”
Grid Long Short-Term Memory (2015, background): “These properties allow LSTM networks to process data with complex and separated interdependencies and to excel in a range of sequence learning domains such as speech recognition [14], offline hand-writing recognition [15], machine translation [35] and image-to-caption generation [36, 25].”
Beyond short snippets: Deep networks for video classification (2015, background): “For this reason, LSTMs yield state-of-the-art results in handwriting recognition [8, 10], speech recognition [9, 7], phoneme detection [5], emotion detection [25], segmentation of meetings and events [18], and evaluating programs [27].”
A Clockwork RNN (2014, methodology): “Its deep variant currently holds the state-of-the-art result in phoneme recognition on the TIMIT database (Graves et al., 2013).”
Dropout Improves Recurrent Neural Networks for Handwriting Recognition (2013, methodology): “This architecture was proposed in [22], but we have adapted the filter sizes for input images at 300 dpi.”
Deep learning in neural networks: An overview (2014, background): “…& Obermayer, 2005), handwriting recognition (Bluche et al., 2014; Graves, Fernandez, Liwicki, Bunke, & Schmidhuber, 2008; Graves et al., 2009; Graves & Schmidhuber, 2009), voice activity detection (Eyben, Weninger, Squartini, & Schuller, 2013), optical character recognition (Breuel,…”
Semantic Image Segmentation with Task-Specific Edge Detection Using CNNs and a Discriminatively Trained Domain Transform (2015, background): “Some recent work has attempted to model spatial long range dependency with recurrent networks for computer vision tasks [17, 41, 35, 4, 43].”
MelNet: A Generative Model for Audio in the Frequency Domain (2019, methodology): “This approach differs from the tightly coupled multidimensional recurrences used by MDRNNs [17, 16] and Grid LSTMs [25] and more closely resembles the approach taken by ReNet [52].”
Self-Delimiting Neural Networks (2012, methodology): “(2) ICDAR 2009 Handwritten Farsi/Arabic Character Recognition Competition.”
Adversarial Generation of Natural Language (2017, background): “, 2014) are powerful models that have been successful at modeling sequential data (Graves and Schmidhuber, 2009; Mikolov et al., 2010).”
Semantic Object Parsing with Graph LSTM (2016, unknown): “LSTM on Image Processing: Recurrent neural networks have been first introduced to address the sequential prediction tasks [24] [25] [26], and then extended to multidimensional image processing tasks [14] [15] such as image generation [2][15], person detection [27], scene labeling [16] and object par”
The Fixed-Size Ordinally-Forgetting Encoding Method for Neural Network Language Models (2015, methodology): “When the 1-of-K representation is used to model a word in a sequence, it can not model its history or context.”
Recurrent convolutional neural network for object recognition (2015, unknown): “In [20] a multi-dimensional RNN (MDRNN) is proposed for off-line handwriting recognition.”
Scene labeling with LSTM recurrent neural networks (2015, methodology): “Image processing task were also considered with the multi-dimensional model [7, 2, 1].”
A Survey on Deep Learning for Multimodal Data Fusion (2020, background): “Unlike deep forward architectures (i.e., DBN, SAE, and CNN), it not only maps the input patterns to the output results but also transfers the hidden states to the outputs by employing the connections between the hidden units (Graves & Schmidhuber, 2008).”