文字的检测与识别资源

持续更新中.......

【综述( Survey)】

[2016Tip] Text Detection Tracking and Recognition in Video [paper]

[2015_PAMI] Text Detection and Recognition in Imagery: A Survey [paper]

[2014_Front.Comput.Sci] Scene Text Detection and Recognition: Recent Advances and Future Trends [paper]

【场景文字检测（Scene Text Detection）】

[2017-CVPR] Detecting Oriented Text in Natural Images by Linking Segments

[2017_AAAI] Detection and Recognition of Text Embedding in Online Images via Neural Context Models [paper] [code]

[2017_AAAI] TextBoxes: A Fast TextDetector with a Single Deep Neural Network [paper][code]

[2016_ECCV] CTPN: Detecting Text in Natural Image with Connectionist Text Proposal Network [paper][demo][code]

[2016_PHD-Thesis]  Context Modeling for Semantic Text Matching and Scene Text Deteciton [paper]

[2016_IJCAI]Scene Text Detection in Video by Learning Locally and Globally [paper]

[201606_arXiv] Scene Text Detection via Holistic, Multi-Channel Prediction [paper]

[2016_CVPR] Accurate Text Localization in Natural Image with Cascaded Convolutional TextNetwork [paper]

[2016_CVPR] Synthetic Data for Text Localization in Natural Images [paper] [data][code]

[2016_CVPR] CannyText Detector: Fast and Robust Scene Text Localization Algorithm [paper]

[2016_CVPR] Multi-oriented text detection with fully convolutional network [paper][code]

[2016_IJCV] Reading Text in the Wild with Convolutional Neural Networks [paper][demo][homepage]

[2016_TIP] Text-Attentional Convolutional Neural Networks for scene Text Detection [paper]

[2016_IJDAR] TextCatcher: a method to detect curved and challenging text in natural scenes [paper]

[2016_arXiv] DeepText: A Unified Framework for Text Proposal Generation and Text Detection in Natural Images [paper][data]

[201601_arXiv] TextProposals: a Text-specific Selective Search Algorithm for Word Spotting in the Wild [paper][code]

[2015_TPAMI] Real-time Lexicon-free Scene Text Localization and Recognition [paper]

[2015_CVPR] Symmetry-Based Text Line Detector in Natural Scenes  [paper][code]

[2015_ICCV] FASText: Efficient unconstrained scene text detector [paper][code]

[2015 ICDAR] Object Proposal for Text Extraction in the Wild [paper][code]

[2015_PHD-Thesis] Deep Learning for Text Spootting [paper]

[2014_ECCV] Deep Features for Text Spotting [paper][code][Homepage]

[2014_TPAMI] Robust Text Detection in Natural Scene Images [paper]

[2014_ECCV] Robust Text Detection with Convolution Neural Network Induced MSER Trees [paper]

[2012_CVPR] Real-time scne text localization and recognition [paper][code]

[2010_CVPR] SWT: Detecting Text in Natural Scenes with Stroke Width Transform [paper] [code]

【自然场景中的文字识别(Scene Text Recognition)】

[2016_NIPS] Generative Shape Models: Joint Text Recognition and Segmentation with Very Little Training Data [paper]

[2016_AAAI] Reading Scene Text in Deep Convolutional Sequences [paper]

[2016_CVPR] Recursive Recurrent Nets with Attention Modeling for OCR in the Wild [paper]

[2016_CVPR] Robust Scene Text Recognition with Automatic Rectification [paper]

[2015_CoRR] An End-to-End Trainable Neural Network for Image-based Sequence Recognition and It's application to Scene Text Recognition [paper][code]

[2015_ICDAR] Automatic Script Identification in the Wild [paper]

[2015_ICLR] Deepstructured output learning for unconstrained text recognition [paper]

[2014_NIPS] Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition [paper] [homepage][model]

[2014_TIP] Aunified Framework for Multi-Oriented Text Detection and Recognition [paper]

[2012_ICPR] End-to-End Text Recognition with CNN [page][code][SVHN DataSet]

【手写体识别(Handwriting Recognition)】

[2016-arXiv] Drawing and Recognizing  Chinese Characters with RNN [paper]

[2016] Learning Spatial-Semantic Context with Fully Convolutional Recurrent Network for Online Handwritten Chinese Text Recognition [paper]

Stroke Sequence-Dependent Deep Convolutional Neural Network for Online HandwrittenChinese Character Recognition [paper]

High Performance Offline Handwritten Chinese Character Recognition Using GoogLeNetand Directional Feature Maps [paper] [github]

DeepHCCR:Offline Handwritten Chinese Character Recognition based on GoogLeNet andAlexNet (With CaffeModel) [code]

Scan,Attend and Read: End-to-End Handwritten Paragraph Recognition with MDLSTMAttention [paper]

【数据集(datasets)】

I. For scene text detection

1. COCO-Text [Homepage][Download]

 63,686images, 173,589 text instances, 3 fine-grained text attributes.

2. Synth-Text [Homepage][Download]

800k thousand images; 8 million synthetic word instances

3. MSRA-TD500

500 natural images that their resolutions of the images vary 1296x864 ~ 1920x1280（300+200）; Chinese, English or mixture of both

4. SVT

 350 high resolution images (average size 1260 × 860) (100 images for training and250 images for testing ) Only word level bounding boxes are provided with case-insensitive labels

5. KAIST

3000 images of indoor and outdoor scenes containing text Korean,English (Number), and Mixed (Korean + English + Number) Task:text location, segmentation and recognition

6. ICDAR系列

-ICDAR 2015 (1000 training images + 500 testing images)

-ICDAR2013 (229 + 233)   [homepage]

-ICDAR2011 (229 + 255)   [Homepage]

-ICDAR2005 (1001 + 489) [Homepage]

-ICDAR2003 (181 + 251)   [Homepage]

II. For Scene Text Recognition

1).  IIIT-5K

5000 images from Scene Texts and born-digital (2k training and 3k testing images)Each image is a cropped word image of scene text with case-insensitive labels

2). Synth-Word

9  million images covering 90k English words (2014 Oxford; VGG)

3) StanfordSynth

Small single-character images of 62 characters (0-9, a-z, A-Z). (2012 Stanford, AI Group)

4) KAIST

5） Chars74K

 Over 74K images from natural images, as well as a set of synthetically generated characters .mall single-character images of 62 characters (0-9, a-z, A-Z).

6) ICDAR系列

【其他(Others)】

I. 开源库

Tesseract: c++ based tools for documents analysis and OCR,support 60+ languages [code]

Ocropy: Python-based tools for document analysis and OCR [code]

CLSTM : A small C++ implementation of LSTM networks,focused on OCR [code]

Convolutional Recurrent Neural Network,Torch7 based [code]

Attention-OCR: Visual Attention based OCR [code]

Umaru: An OCR-system based on torch using the techniqueof LSTM/GRU-RNN, CTC and referred to the works of rnnlib and clstm [code]

2. 相关话题

DeepFont:Identify Your Font from An Image[paper]

Writer-independenFeature Learning for Offline Signature Verification using Deep ConvolutionalNeural Networks[paper]

End-to-EndInterpretation of the French Street Name Signs Dataset [paper] [code]

Extractingtext from an image using Ocropus [blog]

【参考】

[1]http://handong1587.github.io/deep_learning/2015/10/09/ocr.html

[2]https://github.com/chongyangtao/Awesome-Scene-Text-Recognition