Awesome Scene Text

A curated list of papers and resources for scene text detection and recognition

The year when a paper was first published, including ArXiv publications, is used. As a result, there may be cases when a paper was accepted for example to CVPR 2019, but it is listed in year 2018 because it was published in 2018 on ArXiv.

Scene Text Detection (including methods for end-to-end detection and recognition)

2010

• Detecting text in natural scenes with stroke width transform [CVPR 2010] [paper]
• A Method for Text Localization and Recognition in Real-World Images [ACCV 2010] [paper]

2011

2012

• Real-time scene text localization and recognition [CVPR 2012] [paper]

2013

2014

• Robust Scene Text Detection with Convolution Neural Network Induced MSER Trees [ECCV 2014] [paper]

2015

• Symmetry-based text line detection in natural scenes [CVPR 2015] [paper]
• Object proposals for text extraction in the wild [ICDAR 2015] [paper]
• Text-Attentional Convolutional Neural Network for Scene Text Detection [TIP 2016] [paper]
• Text Flow : A Unified Text Detection System in Natural Scene Images [ICCV 2015] [paper]

2016

• Accurate Text Localization in Natural Image with Cascaded Convolutional Text Network [ArXiv] [paper]
• Multi-Oriented Text Detection With Fully Convolutional Networks [CVPR 2016] [paper]
• Scene Text Detection Via Holistic, Multi-Channel Prediction [ArXiv] [paper]
• Detecting Text in Natural Image with Connectionist Text Proposal Network [ECCV 2016] [paper]
  • https://github.com/tianzhi0549/CTPN [Caffe]
  • https://github.com/eragonruan/text-detection-ctpn [TF]
  • https://github.com/Li-Ming-Fan/OCR-DETECTION-CTPN [TF]
• TextBoxes: A Fast Text Detector with a Single Deep Neural Network [AAAI 2017] [paper]
  • https://github.com/MhLiao/TextBoxes [Caffe]
  • https://github.com/shinjayne/shinTB [TF]

2017

• Multi-scale FCN with Cascaded Instance Aware Segmentation for Arbitrary Oriented Word Spotting In The Wild [CVPR 2017] [paper]
• Deep TextSpotter: An End-To-End Trainable Scene Text Localization and Recognition Framework [ICCV 2017] [paper]
• Arbitrary-Oriented Scene Text Detection via Rotation Proposals [TMM 2018] [paper]
  • https://github.com/mjq11302010044/RRPN [Caffe]
• Deep Matching Prior Network: Toward Tighter Multi-oriented Text Detection [CVPR 2017] [paper]
• Detecting Oriented Text in Natural Images by Linking Segments [CVPR 2017] [paper]
  • https://github.com/bgshih/seglink [TF]
  • https://github.com/dengdan/seglink [TF]
• Deep Direct Regression for Multi-Oriented Scene Text Detection [ICCV 2017] [paper]
• Cascaded Segmentation-Detection Networks for Word-Level Text Spotting [ArXiv] [paper]
• EAST: An Efficient and Accurate Scene Text Detector [CVPR 2017] [paper]
  • https://github.com/argman/EAST [TF]
  • https://github.com/kurapan/EAST [Keras]
• WordFence: Text Detection in Natural Images with Border Awareness [ICIP 2017] [paper]
• R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection [ArXiv] [paper]
  • https://github.com/DetectionTeamUCAS/R2CNN_Faster-RCNN_Tensorflow [TF]
  • https://github.com/beacandler/R2CNN [Caffe]
• WordSup: Exploiting Word Annotations for Character based Text Detection [ICCV 2017] [paper]
• Single Shot Text Detector With Regional Attention [ICCV 2017] [paper]
  • https://github.com/BestSonny/SSTD [Caffe]
  • https://github.com/HotaekHan/SSTDNet [PyTorch]
• Fused Text Segmentation Networks for Multi-oriented Scene Text Detection [ArXiv] [paper]
• Deep Residual Text Detection Network for Scene Text [ICDAR 2017] [paper]
• Feature Enhancement Network: A Refined Scene Text Detector [AAAI 2018] [paper]
• ArbiText: Arbitrary-Oriented Text Detection in Unconstrained Scene [ArXiv] [paper]
• Self-organized Text Detection with Minimal Post-processing via Border Learning [ICCV 2017] [paper]
  • https://gitlab.com/rex-yue-wu/ISI-PPT-Text-Detector [Keras]

2018

• PixelLink: Detecting Scene Text via Instance Segmentation [AAAI 2018] [paper]
  • https://github.com/ZJULearning/pixel_link [TF]
  • https://github.com/BowieHsu/tensorflow_ocr [TF]
• FOTS: Fast Oriented Text Spotting With a Unified Network [CVPR 2018] [paper]
• TextBoxes++: A Single-Shot Oriented Scene Text Detector [TIP 2018] [paper]
  • https://github.com/MhLiao/TextBoxes_plusplus [Caffe]
• Multi-oriented Scene Text Detection via Corner Localization and Region Segmentation [CVPR 2018] [paper]
• An end-to-end TextSpotter with Explicit Alignment and Attention [CVPR 2018] [paper]
  • https://github.com/tonghe90/textspotter [Caffe]
• Rotation-Sensitive Regression for Oriented Scene Text Detection [CVPR 2018] [paper]
  • https://github.com/MhLiao/RRD [Caffe]
• Detecting multi-oriented text with corner-based region proposals [Neurocomputing 2019] [paper]
  • https://github.com/xhzdeng/crpn [Caffe]
• An Anchor-Free Region Proposal Network for Faster R-CNN based Text Detection Approaches [ArXiv] [paper]
• IncepText: A New Inception-Text Module with Deformable PSROI Pooling for Multi-Oriented Scene Text Detection [IJCAI 2018] [paper]
  • https://github.com/xieyufei1993/InceptText-Tensorflow [TF]
• Shape Robust Text Detection with Progressive Scale Expansion Network [CVPR 2019] [paper] [paper v2]
  • https://github.com/whai362/PSENet [PyTorch]
  • https://github.com/liuheng92/tensorflow_PSENet [TF]
• TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes [ECCV 2018] [paper]
  • https://github.com/princewang1994/TextSnake.pytorch [PyTorch]
• Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes [ECCV 2018] [paper]
  • https://github.com/lvpengyuan/masktextspotter.caffe2 [Caffe2]
• Accurate Scene Text Detection through Border Semantics Awareness and Bootstrapping [ECCV 2018] [paper]
• A New Anchor-Labeling Method For Oriented Text Detection Using Dense Detection Framework [SPL 2018] [paper]
• An Efficient System for Hazy Scene Text Detection using a Deep CNN and Patch-NMS [ICPR 2018] [paper]
• Scene Text Detection with Supervised Pyramid Context Network [AAAI 2019] [paper]
• Pixel-Anchor: A Fast Oriented Scene Text Detector with Combined Networks [ArXiv] [paper]
• Mask R-CNN with Pyramid Attention Network for Scene Text Detection [WACV 2019] [paper]
• TextMountain: Accurate Scene Text Detection via Instance Segmentation [ArXiv] [paper]
• TextField: Learning A Deep Direction Field for Irregular Scene Text Detection [ArXiv] [paper]
• TextNet: Irregular Text Reading from Images with an End-to-End Trainable Network [ACCV 2018] [paper]

2019

• MSR: Multi-Scale Shape Regression for Scene Text Detection [IJCAI 2019] [paper]
• Scene Text Detection with Inception Text Proposal Generation Module [ICMLC 2019] [paper]
• Towards Robust Curve Text Detection with Conditional Spatial Expansion [CVPR 2019] [paper]
• Curve Text Detection with Local Segmentation Network and Curve Connection [ArXiv] [paper]
• Pyramid Mask Text Detector [ArXiv] [paper]
• Tightness-aware Evaluation Protocol for Scene Text Detection [CVPR 2019] [paper]
• Character Region Awareness for Text Detection [CVPR 2019] [paper]
• Look More Than Once: An Accurate Detector for Text of Arbitrary Shapes [CVPR 2019] [paper]
• TextCohesion: Detecting Text for Arbitrary Shapes [ArXiv] [paper]
• Arbitrary Shape Scene Text Detection With Adaptive Text Region Representation [CVPR 2019] [paper]
• Learning Shape-Aware Embedding for Scene Text Detection [CVPR 2019] [paper]
• A Single-Shot Arbitrarily-Shaped Text Detector based on Context Attended Multi-Task Learning [ACMMM 2019] [paper]
• Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network [ICCV 2019] [paper]
• Towards Unconstrained End-to-End Text Spotting [ICCV 2019] [paper]
• TextDragon: An End-to-End Framework for Arbitrary Shaped Text Spotting [paper]
• Convolutional Character Networks [ICCV 2019] [paper]

Weakly supervised Scene Text Detection & Recognition

2017

• Attention-Based Extraction of Structured Information from Street View Imagery [ICDAR 2017] [paper]
• WeText: Scene Text Detection under Weak Supervision [ICCV 2017] [paper]
• SEE: Towards Semi-Supervised End-to-End Scene Text Recognition [AAAI 2018] [paper]
  • https://github.com/Bartzi/see [Chainer]

Scene Text Recognition

2014

• Deep Structured Output Learning for Unconstrained Text Recognition [ICLR 2015] [paper]
  • https://github.com/AlexandreSev/Structured_Data [TF]
• Reading text in the wild with convolutional neural networks [IJCV 2016] [paper]
  • https://github.com/mathDR/reading-text-in-the-wild [Keras]

2015

• Reading Scene Text in Deep Convolutional Sequences [AAAI 2016] [paper]
• An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition [TPAMI 2017] [paper]
  • https://github.com/bgshih/crnn [Torch]
  • https://github.com/weinman/cnn_lstm_ctc_ocr [TF]
  • https://github.com/watsonyanghx/CNN_LSTM_CTC_Tensorflow [TF]
  • https://github.com/MaybeShewill-CV/CRNN_Tensorflow [TF]
  • https://github.com/meijieru/crnn.pytorch [PyTorch]
  • https://github.com/kurapan/CRNN [Keras]

2016

• Recursive Recurrent Nets with Attention Modeling for OCR in the Wild [CVPR 2016] [paper]
• Robust scene text recognition with automatic rectification [CVPR 2016] [paper]
  • https://github.com/WarBean/tps_stn_pytorch [PyTorch]
  • https://github.com/marvis/ocr_attention [PyTorch]
• CNN-N-Gram for Handwriting Word Recognition [CVPR 2016] [paper]
• STAR-Net: A SpaTial Attention Residue Network for Scene Text Recognition [BMVC 2016] [paper]

2017

• STN-OCR: A single Neural Network for Text Detection and Text Recognition [ArXiv] [paper]
  • https://github.com/Bartzi/stn-ocr [MXNet]
• Learning to Read Irregular Text with Attention Mechanisms [IJCAI 2017] [paper]
• Scene Text Recognition with Sliding Convolutional Character Models [ArXiv] [paper]
• Focusing Attention: Towards Accurate Text Recognition in Natural Images [ICCV 2017] [paper]
• AON: Towards Arbitrarily-Oriented Text Recognition [CVPR 2018] [paper]
  • https://github.com/huizhang0110/AON [TF]
• Gated Recurrent Convolution Neural Network for OCR [NIPS 2017] [paper]
  • https://github.com/Jianfeng1991/GRCNN-for-OCR [Torch]

2018

• Char-Net: A Character-Aware Neural Network for Distorted Scene Text Recognition [AAAI 2018] [paper]
• SqueezedText: A Real-time Scene Text Recognition by Binary Convolutional Encoder-decoder Network [AAAI 2018] [paper]
• Edit Probability for Scene Text Recognition [CVPR 2018] [paper]
• ASTER: An Attentional Scene Text Recognizer with Flexible Rectification [TPAMI 2018] [paper]
  • https://github.com/bgshih/aster [TF]
• Synthetically Supervised Feature Learning for Scene Text Recognition [ECCV 2018] [paper]
• Scene Text Recognition from Two-Dimensional Perspective [AAAI 2019] [paper]
• ESIR: End-to-end Scene Text Recognition via Iterative Image Rectification [CVPR 2019] [paper]

2019

• A Multi-Object Rectified Attention Network for Scene Text Recognition [Pattern Recognition] [paper]
  • https://github.com/Canjie-Luo/MORAN_v2 [PyTorch]
• A Simple and Robust Convolutional-Attention Network for Irregular Text Recognition [paper]
• Aggregation Cross-Entropy for Sequence Recognition [CVPR 2019][paper]
  • https://github.com/summerlvsong/Aggregation-Cross-Entropy [PyTorch]
• Sequence-to-Sequence Domain Adaptation Network for Robust Text Image Recognition [CVPR 2019][paper]
• 2D Attentional Irregular Scene Text Recognizer [ArXiv] [paper]
• Deep Neural Network for Semantic-based Text Recognition in Images [ArXiv] [paper]
• Symmetry-constrained Rectification Network for Scene Text Recognition [ICCV 2019] [paper]
• Rethinking Irregular Scene Text Recognition (ICDAR 2019-ArT) [paper]
  • https://github.com/Jyouhou/ICDAR2019-ArT-Recognition-Alchemy [PyTorch]
• Focus-Enhanced Scene Text Recognition with Deformable Convolutions [ArXiv] [paper]
  • https://github.com/Alpaca07/dtr [PyTorch]
• Adaptive Embedding Gate for Attention-Based Scene Text Recognition [ArXiv] [paper]

Script Identification

Other scene text related papers

2016

• Synthetic Data for Text Localisation in Natural Images [CVPR 2016] [paper]
  • https://github.com/ankush-me/SynthText

2019

• Scene Text Synthesis for Efficient and Effective Deep Network Training [ArXiv] [paper]

Scene text survey

2018

• Scene Text Detection and Recognition: The Deep Learning Era [ArXiv] [paper]

2019

• Scene text detection and recognition with advances in deep learning: a survey [IJDAR 2019] [paper]