ICDAR2017 Competition on Handwritten Text Recognition on the READ Dataset

Lafuente, José Vicente; Romero, Verónica; Toselli, Alejandro H.; Villegas, Mauricio; Vidal, Enrique

doi:10.1109/icdar.2017.226

Cited by 42 publications

(24 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ICDAR2017 full page HTR competition [24] consists of two training sets. The first contains 50 fully annotated images with line-level localization and transcription ground-truth.…”

Section: Datasetsmentioning

confidence: 99%

“…However, line segmentation is still an error-prone process and can cause great deterioration in the performance of today's text recognition systems. This is especially true for documents with hard to segment text-lines such as handwritten documents [10,24], with warped lines, uneven interline spacing, touching lines, and torn pages.…”

Section: Introductionmentioning

confidence: 99%

“…On the challenging ICDAR 2017 HTR [24] full page benchmark we achieve state-of-the-art Character Error Rate (CER) without any localization data. On full paragraphs of the IAM [17] dataset, we were able to achieve state-ofthe-art CER surpassing models that work on carefully presegmented text-lines, without using any localization infor-mation during training or testing.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

OrigamiNet: Weakly-Supervised, Segmentation-Free, One-Step, Full Page Text Recognition by learning to unfold

Yousef¹,

Bishop²

2020

Preprint

View full text Add to dashboard Cite

Text recognition is a major computer vision task with a big set of associated challenges. One of those traditional challenges is the coupled nature of text recognition and segmentation. This problem has been progressively solved over the past decades, going from segmentation based recognition to segmentation free approaches, which proved more accurate and much cheaper to annotate data for. We take a step from segmentation-free single line recognition towards segmentation-free multi-line / full page recognition. We propose a novel and simple neural network module, termed OrigamiNet, that can augment any CTC-trained, fully convolutional single line text recognizer, to convert it into a multi-line version by providing the model with enough spatial capacity to be able to properly collapse a 2D input signal into 1D without losing information. Such modified networks can be trained using exactly their same simple original procedure, and using only unsegmented image and text pairs. We carry out a set of interpretability experiments that show that our trained models learn an accurate implicit line segmentation. We achieve state-of-the-art character error rate on both IAM & ICDAR 2017 HTR benchmarks for handwriting recognition, surpassing all other methods in the literature. On IAM we even surpass single line methods that use accurate localization information during training. Our code is available online at https: //github.com/IntuitionMachines/OrigamiNet.

show abstract

“…The ICDAR2017 full page HTR competition [24] consists of two training sets. The first contains 50 fully annotated images with line-level localization and transcription ground-truth.…”

Section: Datasetsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

OrigamiNet: Weakly-Supervised, Segmentation-Free, One-Step, Full Page Text Recognition by learning to unfold

Yousef¹,

Bishop²

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…In particular, Graves [11] presented a neural network based on interleaved convolutional and 2D-LSTM (Long-Short Term Memory [13]) layers that were trained using the Connectionist Temporal Classification (CTC) strategy [10]. This pioneering approach yielded good results on various datasets in several languages [17] and most of major recent competitions were won by systems with related neural network architectures [1,17,23,[27][28][29].…”

Section: Introductionmentioning

confidence: 99%

Are 2D-LSTM really dead for offline text recognition?

Moysset

Messina

2019

IJDAR

View full text Add to dashboard Cite

There is a recent trend in handwritten text recognition with deep neural networks to replace 2D recurrent layers with 1D, and in some cases even completely remove the recurrent layers, relying on simple feed-forward convolutional only architectures. The most used type of recurrent layer is the Long-Short Term Memory (LSTM). The motivations to do so are many:there are few open-source implementations of 2D-LSTM, even fewer supporting GPU implementations (currently cuDNN only implements 1D-LSTM); 2D recurrences reduce the amount of computations that can be parallelized, and thus possibly increase the training/inference time; recurrences create global dependencies with respect to the input, and sometimes this may not be desirable.Many recent competitions were won by systems that employed networks that use 2D-LSTM layers. Most previous work that compared 1D or pure feed-forward architectures to 2D recurrent models have done so on simple datasets or did not fully optimize the "baseline" 2D model compared to the challenger model, which was dully optimized.In this work, we aim at a fair comparison between 2D and competing models and also extensively evaluate them on more complex datasets that are more representative of challenging "real-world" data, compared to "academic" datasets that are more restricted in their complexity. We aim at determining when and why the 1D and 2D recurrent models have different results. We also compare the results with a language model to as-Bastien Moysset A2iA SA, sess if linguistic constraints do level the performance of the different networks.Our results show that for challenging datasets, 2D-LSTM networks still seem to provide the highest performances and we propose a visualization strategy to explain it.

show abstract

“…Therefore, new projects involve the use of several domains such as language processing, document recognition, and information retrieval for historical studies. In recent years, the number of competitions regarding historical documents has increased [1], [2]. Such systems have to deal with the complexity of the task as well as the document medium, its level of deterioration, or even its written language.…”

Section: Introductionmentioning

confidence: 99%

Separating Optical and Language Models Through Encoder-Decoder Strategy for Transferable Handwriting Recognition

Granet¹,

Morin²,

Mouchère³

et al. 2018

2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR)

View full text Add to dashboard Cite

Lack of data can be an issue when beginning a new study on historical handwritten documents. To deal with this, we propose a deep-learning based recognizer which separates the optical and the language models in order to train them separately using different resources. In this work, we present the optical encoder part of a multilingual transductive transfer learning applied to historical handwriting recognition. The optical encoder transforms the input word image into a non-latent space that depends only on the letter-n-grams: it enables it to be independent of the language. This transformation avoids embedding a language model and operating the transfer learning across languages using the same alphabet. The language decoder creates from a vector of letter-n-grams a word as a sequence of characters. Experiments show that separating optical and language model can be a solution for multilingual transfer learning.

show abstract

ICDAR2017 Competition on Handwritten Text Recognition on the READ Dataset

Cited by 42 publications

References 3 publications

OrigamiNet: Weakly-Supervised, Segmentation-Free, One-Step, Full Page Text Recognition by learning to unfold

OrigamiNet: Weakly-Supervised, Segmentation-Free, One-Step, Full Page Text Recognition by learning to unfold

Are 2D-LSTM really dead for offline text recognition?

Separating Optical and Language Models Through Encoder-Decoder Strategy for Transferable Handwriting Recognition

Contact Info

Product

Resources

About