E2E-MLT - An Unconstrained End-to-End Method for Multi-language Scene Text

Bušta, Michal; Patel, Yash; Matas, Jiřı́

doi:10.1007/978-3-030-21074-8_11

Cited by 58 publications

(56 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, we compare our network with the stateof-the-art approaches [1], [3], [10], [11], [16], [18], [20], [21], [23], [24], [27]- [29], [43], [45], [49], [65], [66], [69], [69], [71]- [73] on six different benchmark datasets. We consider recall, precision, and f-measure as the metrics for evaluation of accuracy of detection.…”

Section: Comparison With State-of-the-art Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Cluttered TextSpotter: An End-to-End Trainable Light-Weight Scene Text Spotter for Cluttered Environment

2020

View full text Add to dashboard Cite

Scene text spotting aims at simultaneously localizing and recognizing text instances, symbols, and logos in natural scene images. Scene text detection and recognition approaches have received immense attention in computer vision research community. The presence of partial occlusion or truncation artifact due to the cluttered background of scene images creates an obstacle in perceiving the text instances, which makes the process of spotting very complex. In this paper, we propose a lightweight scene text spotter that can address the issue of cluttered environment of scene images. It is an end-to-end trainable deep neural network that uses local part information, global structural features, and context cue information of oriented region proposals for spotting text instances. It helps to localize in scene images with background clutters, where partially occluded text parts, truncation artifacts, and perspective distortions are present. We mitigate the problem of misclassification caused by inter-class interference by exploring inter-class separability and intra-class compactness. We also incorporate multi-language character segmentation and word-level recognition in a lightweight recognition module. We have used six publicly available benchmark datasets in different smart devices to illustrate the efficacy of the network.

show abstract

Section: Comparison With State-of-the-art Resultsmentioning

confidence: 99%

“…Each proposal uses a connectionist temporal classification to decode multi-language texts. E2E-MLT [71] is one of the popular multilingual optical character recognition for scene text detection and recognition. It uses a single shared fully convolutional network.…”

Section: Scene Text Spottingmentioning

confidence: 99%

Cluttered TextSpotter: An End-to-End Trainable Light-Weight Scene Text Spotter for Cluttered Environment

2020

View full text Add to dashboard Cite

show abstract

“…They reported a character recognition rate of 98.17%, 97.44% for the Arabic news video text and 75.05% for the synthetic Arabic natural scene character image datasets. A multi-language end-to-end scene text recognition system was presented in [58]. The authors used the ResNet50 [59] network for text localization, VGG16 [60] pre-trained on ImageNet [61] data for script identification and an OCR module proposed in [62] for multi-language text recognition.…”

Section: Ieee Accessmentioning

confidence: 99%

Cursive Character Recognition in Natural Scene Images Using a Multilevel Convolutional Neural Network Fusion

2020

View full text Add to dashboard Cite

The accuracy of current natural scene text recognition algorithms is limited by the poor performance of character recognition methods for these images. The complex backgrounds, variations in the writing, text size, orientations, low resolution and multi-language text make recognition of text in natural images a complex and challenging task. Conventional machine learning and deep learning-based methods have been developed that have achieved satisfactory results, but character recognition for cursive text such as Arabic and Urdu scripts in natural images is still an open research problem. The characters in the cursive text are connected and are difficult to segment for recognition. Variations in the shape of a character due to its different positions within a word make the recognition task more challenging than non-cursive text. Optical character recognition (OCR) techniques proposed for Arabic and Urdu scanned documents perform very poorly when applied to character recognition in natural images. In this paper, we propose a multiscale feature aggregation (MSFA) and a multi-level feature fusion (MLFF) network architecture to recognize isolated Urdu characters in natural images. The network first aggregates multi-scale features of the convolutional layers by up-sampling and addition operations and then combines them with the high-level features. Finally, the outputs of the MSFA and MLFF networks are fused together to create more robust and powerful features. A comprehensive dataset of segmented Urdu characters is developed for the evaluation of the proposed network models. Synthetic text on the patches of images with real natural scene backgrounds is generated to increase the samples of infrequently used characters. The proposed model is evaluated on the Chars74K and ICDAR03 datasets. To validate the proposed model on the new Urdu character image dataset, we compare its performance with the histogram of oriented gradients (HoG) method. The experimental results show that the aggregation of multi-scale and multilevel features and their fusion is more effective, and outperforms other methods on the Urdu character image and Chars74K datasets. INDEX TERMS Cursive text recognition, natural scene Urdu character recognition, multi-scale feature aggregation, multi-level feature fusion, convolutional neural network (CNN)

show abstract

“…In recent years, with the renaissance of convolutional neural networks (CNNs), many deep learning-based methods [ 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 ] have achieved remarkable achievements in text detection, and these methods can be divided into top-down and bottom-up methods. The top-down methods [ 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 ], also commonly referred to as regression-based methods, usually adopt popular object detection pipelines to first detect text on the block level and then break a block into the word or line level if necessary. However, because of the structural limitations of the corresponding CNN models, these methods cannot efficiently handle long text and arbitrari...…”

Section: Introductionmentioning

confidence: 99%

A Straightforward and Efficient Instance-Aware Curved Text Detector

Zhao

Shao

Zhang

et al. 2021

Sensors

View full text Add to dashboard Cite

A challenging aspect of scene text detection is to handle curved texts. In order to avoid the tedious manual annotations for training curve text detector, and to overcome the limitation of regression-based text detectors to irregular text, we introduce straightforward and efficient instance-aware curved scene text detector, namely, look more than twice (LOMT), which makes the regression-based text detection results gradually change from loosely bounded box to compact polygon. LOMT mainly composes of curve text shape approximation module and component merging network. The shape approximation module uses a particle swarm optimization-based text shape approximation method (called PSO-TSA) to fine-tune the quadrilateral text detection results to fit the curved text. The component merging network merges incomplete text sub-parts of text instances into more complete polygon through instance awareness, called ICMN. Experiments on five text datasets demonstrate that our method not only achieves excellent performance but also has relatively high speed. Ablation experiments show that PSO-TSA can solve the text’s shape optimization problem efficiently, and ICMN has a satisfactory merger effect.

show abstract

E2E-MLT - An Unconstrained End-to-End Method for Multi-language Scene Text

Cited by 58 publications

References 46 publications

Cluttered TextSpotter: An End-to-End Trainable Light-Weight Scene Text Spotter for Cluttered Environment

Cluttered TextSpotter: An End-to-End Trainable Light-Weight Scene Text Spotter for Cluttered Environment

Cursive Character Recognition in Natural Scene Images Using a Multilevel Convolutional Neural Network Fusion

A Straightforward and Efficient Instance-Aware Curved Text Detector

Contact Info

Product

Resources

About