A Method for Text Localization and Recognition in Real-World Images

By Lukás Neumann, Jiri Matas. 2010

592 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

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

1. Xiang Bai
2. Luc Van Gool
3. Andrew Zisserman
4. Cong Yao
5. Andrea Vedaldi
6. Chunhua Shen
7. Dacheng Tao
8. Yu Qiao

Citations in the corpus

• Speeding up Convolutional Neural Networks with Low Rank Expansions (2014, background): “Character classification is an essential part of many text spotting pipelines such as [3, 4, 15, 24, 25, 26, 27, 29, 30, 41, 43].”
• Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition (2014, methodology): “Traditional text recognition methods are based on sequential character classification by either sliding windows [11, 26, 27] or connected components [18, 19], after which a word prediction is made by grouping character classifier predictions in a left-to-right manner.”
• Reading Text in the Wild with Convolutional Neural Networks (2014, methodology): “3.1 Text Detection Methods Text detection methods tackle the first task of the standard text spotting pipeline [11]: producing segmentations or bounding boxes of words in natural scene images.”
• End-to-End Text Recognition with Hybrid HMM Maxout Models (2013, methodology): “The works of (Neumann & Matas, 2011; Mishra et al., 2012; Novikova et al., 2012) are the only works we know of that show how their meth-ods scale with lexicon size.”
• Cascaded Segmentation-Detection Networks for Text-Based Traffic Sign Detection (2018, background): “Neumann and Matas [35] extract MSERs as candidates to detect texts.”
• EAST: An Efficient and Accurate Scene Text Detector (2017, methodology): “Numerous inspiring ideas and effective approaches [5, 25, 26, 24, 27, 37, 11, 12, 7, 41, 42, 31] have been investigated.”
• Scene Text Localization and Recognition with Oriented Stroke Detection (2013, methodology): “The second, recently more popular approach [4, 13, 11, 19, 12, 15] is based on localizing individual characters as connected components using local properties of an image (color, intensity, stroke-width, etc.”
• End-to-end text recognition with convolutional neural networks (2012, background): “40 obtained with a general lexicon by [14].”
• Top-down and bottom-up cues for scene text recognition (2012, background): “Standard off-the-shelf OCRs perform very poorly on these datasets [23, 29].”
• Real-time scene text localization and recognition (2012, methodology): “The proposed method differs from the MSER-based methods [14, 15] in that it tests all ERs (not only the subset of MSERs) while reducing the memory footprint and maintaining the same computational complexity and real-time performance.”
• TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes (2018, methodology): “In the past few years, the most prominent trend in the area of scene text detection is the transfer from conventional methods [14,15] to deep learning based methods [16,17,4,3,2].”
• A Fast Uyghur Text Detector for Complex Background Images (2018, background): “extracted as candidate components in [7], [33], [34].”
• PhotoOCR: Reading Text in Uncontrolled Conditions (2013, methodology): “For example the method of [18] uses a bigram language model together with a set of hand-designed image features and a support vector machine classifier for text detection and recognition.”
• A Unified Framework for Multioriented Text Detection and Recognition (2014, methodology): “To enable fair comparison, we follow the evaluation protocol of Neumann et al. [12].”
• Arbitrary-Oriented Scene Text Detection via Rotation Proposals (2017, background): “In the past years there have been many attentions in the text detection task [3, 22, 7, 35, 23, 1, 15, 12, 31].”
• Deep Features for Text Spotting (2014, methodology): “Text detection methods are either based on connected components (CCs) [11, 16, 50, 49, 32–35] or sliding windows [40, 7, 39, 45].”
• Recognizing Text with Perspective Distortion in Natural Scenes (2013, methodology): “The typical features used for detection are stroke widths [9, 10], Maximally Stable Extremal Regions (MSER) [11, 12] and HOG [1].”
• Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes (2018, background): “A large number of methods have been proposed to detect scene text [7, 36, 37, 50, 19, 23, 54, 21, 47, 54, 56, 30, 52, 55, 34, 15, 48, 43, 57, 16, 35, 31].”
• Integrating Scene Text and Visual Appearance for Fine-Grained Image Classification (2017, methodology): “In the detection stage, connected components [38], [40], [41] or sliding windows [39], [42] are usually used to extract character candidates.”
• Large-Lexicon Attribute-Consistent Text Recognition in Natural Images (2012, methodology): “[3, 4, 11] use maximally-stable extremal regions (MSER) as character candidates and achieve a high recall.”