Text Localization in Real-World Images Using Efficiently Pruned Exhaustive Search

By Lukás Neumann, Jiri Matas. 2011

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

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].”
• Reading Text in the Wild with Convolutional Neural Networks (2014, neutral): “Neumann [42] - - 41 - - - - Wang [56] 68 51 - - 38 - - - Wang & Wu [57] 72 67 - - 46 - - - Neumann [44] - - - - - - 45 Alsharif [5] 77 70 63 - 48 - - - Jaderberg [31] 80 75 - - 56 - - - Proposed 90 86 78 72 76 53 76 69 76 Proposed (0.”
• 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.”
• A Fast Uyghur Text Detector for Complex Background Images (2018, methodology): “In particular, FASTroke compared with I-MSER, obtains a higher recall, but the detected components have reduced at least by half.”
• Scene text detection and recognition: recent advances and future trends (2015, methodology): “[10, 65] to attain real-time text detection and recognition2) (Fig.”
• Text-Attentional Convolutional Neural Network for Scene Text Detection (2015, methodology): “The main difficulties for this group of methods lie in designing a discriminative feature to train a powerful classifier, and managing computational flexibility by reducing the number of the scanning windows.”
• Deep Features for Text Spotting (2014, methodology): “For example, Epshtein et al. take characters as CCs of the stroke width transform [16], while Neumann and Matas [34, 33] use Extremal Regions [29], or more recently oriented strokes [35], as CCs representing characters.”
• TextProposals: A text-specific selective search algorithm for word spotting in the wild (2016, methodology): “Neumann and Matas [33] have introduced an efficient exhaustive search algorithm using heuristic verification functions at different grouping levels (i.”
• Learning to Detect Partially Overlapping Instances (2013, methodology): “Such detection can be based on a sliding window or Hough transform, followed by an appropriate non-maxima suppression procedure [3, 8, 14], stochastic fitting of interacting particles or object models [9, 10, 24], or region-based detection [2, 18, 19].”
• Real-time scene text localization and recognition (2012, methodology): “A highly efficient exhaustive search with feedback loops (adapted from [15]) is then applied to group ERs into words and select the most probable character segmentation.”