papers.md

Scene Text Recognition Papers

Contents

• Methods
  • CTC
  • Attention
  • Transformer
  • Rectification
  • Language Model
  • Dataset
  • Data Augmentation
  • Survey
  • Self Supervise
  • Super Resolution
  • Diffusion Models
  • Others
• Conference
  • CVPR
  • ICCV
  • ECCV
  • AAAI
  • NIPS
  • others
• Journal
  • TPAMI
  • PR

Methods

CTC

• Pattern Recognition-2020,引用数:3:Reinterpreting CTC training as iterative fitting

  • 探讨CTC数学原理，将CTC Loss解释为交叉熵损失，较为理论

• ECCV-2020,引用数:2:Variational Connectionist Temporal Classification

  • 提出变分CTC来增强网络对于非blank符号的学习

• AAAI-2020,引用数:27:GTC: Guided Training of CTC towards Efficient and Accurate Scene Text Recognition

  • 训练的时候用Attention分支辅助CTC，测试的时候只用CTC

• IEEE Access-2019,引用数:30:Natural Scene Text Recognition Based on Encoder-Decoder Framework

  • Attention的解码时候的对齐是没有限制的，故引入CTC对Attention的对齐进行监督

• ECCV-2018, 引用数:69:Synthetically supervised feature learning for scene text recognition

• NIPS-2018,引用数:25:Connectionist Temporal Classification with Maximum Entropy Regularization

  • 解决CTC中的Spiky Distribution Problem, 使用最大熵来限制CTC学习，较为理论

• NIPS-2017, 引用数:105:Gated recurrent convolution neural network for OCR

  • GRCNN

• Pattern Recognition-2017, 引用数:99:Accurate recognition of words in scenes without character segmentation using recurrent neural network

• BMVC-2016,引用数:136:STAR-Net: A spatial attention residue network for scene text recognition

• TPAMI-2016,引用数:1497:An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition

  • 场景文字识别开山之作，引入CTC将识别变为一个序列到序列的问题。

---

Attention

• IJCAI-2022:SVTR: Scene Text Recognition with a Single Visual Model
• ICDAR2021:Representation and Correlation Enhanced Encoder-Decoder Framework for Scene Text Recognition
• Electronics 2021: TRIG: Transformer-Based Text Recognizer with Initial Embedding Guidance
• Patter Recognition-2021,引用数:23:Master: Multi-aspect non-local network for scene text recognition
• ECCV-2020, 引用数:27:Robustscanner: Dynamically enhancing positional clues for robust text recognition
• CVPR-2020, 引用数:42:SCATTER: selective context attentional scene text recognizer
  • 多阶段
• CVPRWorkshop-2020, 引用数:28:On Recognizing Texts of Arbitrary Shapes with 2D Self-Attention
  • Transformer encoder + decoder, 引入自适应的2D位置编码，对于旋转文字，多行文字有较好的鲁棒性
• AAAI-2020, 引用数：28:Textscanner: Reading characters in order for robust scene text recognition
  • 实例分割
• AAAI-2020, 引用数：67:Decoupled attention network for text recognition
  • 考虑到注意力机制对于长文本的飘移，采用UNet架构来直接生成注意力图，将注意力图和解码之间解耦开
• Neural Computing-2020, 引用数:17:Adaptive embedding gate for attention-based scene text recognition
• ICCV-2019, 引用数:204:What is wrong with scene text recognition model comparisons? dataset and model analysis
  • 框架型文章，值得一读
• AAAI-2019，引用数:128:Show, attend and read: A simple and strong baseline for irregular text recognition
  • 2D Attention
  • 在Attention计算过程中引入门控机制
• ICDAR-2019, 引用数:45NRTR: A No-Recurrence Sequence-to-Sequence Model For Scene Text Recognition
  • 使用Transformer encoder和decoder
• ACM MM-2018, 引用数:44:Attention and language ensemble for scene text recognition with convolutional sequence modeling
• CVPR-2018, 引用数:93:Edit probability for scene text recognition
  • 现有的attention方法采用最大似然损失函数，本文探讨输出概率分布和预测间的关系
• AAAI-2018, 引用数:103:Char-Net: A character-aware neural network for distorted scene text recognition
  • 加入单字检测分支。
• AAAI-2018, 引用数:50:SEE: Towards Semi-Supervised End-to-End Scene Text Recognition
  • 半监督端到端识别。
• TPAMI-2018, 引用数:329:ASTER: An Attentional Scene Text Recognizer with Flexible Rectification
  • Attention+Rectification经典之作
• Neural Computing-2018, 引用数:41:Reading scene text with fully convolutional sequence modeling
  • Attention算法使用RNN建模，计算复杂并且较难训练，本文使用全卷积网络来捕获全局信息，比BiLSTM更加有效
• CVPR-2018,引用数:196:AON: Towards arbitrarily-oriented text recognition
  • 关注于不规则文字的识别
• ICCV-2017, 引用数:290:Focusing attention: Towards accurate text recognition in natural images
  • attention存在注意力飘移问题，引入Focusing Network把飘移的注意力抓回来
• IJCAI-2017, 引用数:124:Learning to read irregular text with attention mechanisms
• CVPR-2016, 引用数:370:Recursive recurrent nets with attention modeling for OCR in the wild

---

Transformer

• ECCV-2022: Pure Transformer with Integrated Experts for Scene Text Recognition
• BMCV-2022: Masked Vision-Language Transformers for Scene Text Recognition

---

Rectification Model

• BMCV-2021:An Adaptive Rectification Model for Arbitrary-Shaped Scene Text Recognition

  • 提出新的矫正方法，在弯曲文本上效果好于TPS和MORAN

• ICCV-2019, 引用数:77:Symmetry-constrained Rectification Network for Scene Text Recognition

  • 带限制的矫正网络

• CVPR-2019, 引用数:165:ESIR: End-to-end scene text recognition via iterative image rectification

  • 迭代矫正

• TPAMI-2018, 引用数:329:ASTER: An Attentional Scene Text Recognizer with Flexible Rectification

  • 引入TPS变换进行矫正

• Pattern Recognition-2018, 引用数:161MORAN: A Multi-Object Rectified Attention Network for Scene Text Recognition

  • 任意方向矫正，效果比ASTER出色

• CVPR-2016, 引用数:415:Robust Scene Text Recognition With Automatic Rectification

  • TPS矫正+Attention

---

Language Model

• BMVC-2022:Visual-semantic transformer for scene text recognition
• BMVC-2022:Parallel and Robust Text Rectifier for Scene Text Recognition
• ICFHR-2022:A Vision Transformer Based Scene Text Recognizer with Multi-grained Encoding and Decoding
• TIP-2022:PETR: Rethinking the Capability of Transformer-Based Language Model in Scene Text Recognition
• ACCESS-2022:Scene Text Recognition with Semantics
• AAAI-2022:Visual Semantics Allow for Textual Reasoning Better in Scene Text Recognition
• ECCV-2022:Levenshtein OCR
• ECCV-2022:Multi-Granularity Prediction for Scene Text Recognition
• ECCV-2022:Scene Text Recognition with Permuted Autoregressive Sequence Models
• AAAI-2022:Visual Semantics Allow for Textual Reasoning Better in Scene Text Recognition
  • 提升语言模型对于任意现状文本的识别能力
• arXiv-2021/12/1:Visual-Semantic Transformer for Scene Text Recognition
• arXiv-2021/11/30: Multi-modal Text Recognition Networks: Interactive Enhancements between Visual and Semantic Features
  • 探讨语言模型和视觉模型如何更好的结合，比肩ABINet，取得SOTA
• ICCV-2021，引用数:1 From Two to One: A New Scene Text Recognizer with Visual Language Modeling Network
  • VisionLAN
  • 提出了一个新的遮挡文字数据集
  • 弱监督的将语言模型融入进视觉模型中
• ICCV-2021，引用数:1 Joint Visual Semantic Reasoning: Multi-Stage Decoder for Text Recognition
  • 多阶段+transformer识别器
  • 引入gumbel softmax，解决视觉到语义不可导问题
• CVPR-2021，Oral，引用数:1 Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognition
  • ABINet
  • 屠榜作品，在SRN基础进行改良，从人类阅读的角度进行思考
• CVPR-2020, 引用数:58 Towards accurate scene text recognition with semantic reasoning networks
  • SRN
  • 引入Transformer进行语言建模，视觉+语言模型，取得SOTA效果
• CVPR-2020，引用数:59 Seed: Semantics enhanced encoder-decoder framework for scene text recognition
  • SEED
  • 在BiLSTM第一个单元前输入语义信息
  • 首次尝试把语言模型引入场景文字识别中

---

Dataset

• CVPR-2020, 引用数:21:UnrealText: Synthesizing realistic scene text images from the unreal world

  • 虚幻引擎来渲染文字图像

• CVPR-2016, 引用数:979:Synthetic data for text localisation in natural images

  • SynthText数据集

• NIPS-2014, 引用数:737:Synthetic data and artificial neural net?works for natural scene text recognition
  • MJ数据集

---

Data Augmentation

• Arxiv-2021/11/17 TextAdaIN: Paying Attention to Shortcut Learning in Text Recognizers
  • 使用Permuted AdaIN进行数据增广
• CVPR-2020，引用数:32 Learn to Augment: Joint Data Augmentation and Network Optimization for Text Recognition
  • 利用仿射变化对图像进行数据增广，有效提升准确率

---

Survey

• IJCV-2021, 引用数:156:Scene text detection and recognition: The deep learning era
• ACM Computing Surveys-2020, 引用数:28: Text Recognition in the Wild: A Survey
• TPAMI-2015, 引用数:682: Text detection and recognition in imagery: A survey
• Frontiers of Computer Science-2016, 引用数:299: Scene text detection and recognition: Recent advances and future trends

---

Self Supervise

• ICCV-2023: Self-supervised Character-to-Character Distillation for Text Recognition
• ICCV-2023: Revisiting Scene Text Recognition: A Data Perspective
• WACV-2023:Seq-UPS: Sequential Uncertainty-aware Pseudo-label Selection for Semi-Supervised Text Recognition
• AAAI-2022: Perceiving Stroke-Semantic Context: Hierarchical Contrastive Learning for Robust Scene Text Recognition
• WACV-2023:Seq-UPS: Sequential Uncertainty-aware Pseudo-label Selection for Semi-Supervised Text Recognition
• arXiv-2022:MaskOCR: Text Recognition with Masked Encoder-Decoder Pretraining
• ACM-MM22:Reading and Writing: Discriminative and Generative Modeling for Self-Supervised Text Recognition
• arXiv-2022:Multimodal Semi-Supervised Learning for Text Recognition
• CVPR-2022:SimAN: Exploring Self-Supervised Representation Learning of Scene Text via Similarity-Aware Normalization
  • 生成式对比学习
• AAAI-2022:Context-based Contrastive Learning for Scene Text Recognition
  • 对比学习用于场景文字识别
• CVPR-2021, 引用数:5:Sequence-to-Sequence Contrastive Learning for Text Recognition
  • 首次在STR中引入对比学习的方法

---

Super Resolution

• ICCV-2023: A Benchmark for Chinese-English Scene Text Image Super-resolution

• TIP-2023:Text prior guided scene text image super-resolution

• CVPR-2022:A Text Attention Network for Spatial Deformation Robust Scene Text Image Super-resolution

• AAAI-2022:Text Gestalt: Stroke-Aware Scene Text Image Super-Resolution

  • 场景文本超分

• arXiv-2021/12/16:TRIG: Transformer-Based Text Recognizer with Initial Embedding Guidance

  • TPS + Transformer Encoder + Attention Decoder的组合。
  • 场景文字超分，引入笔画级别的监督

• CVPR2021:Scene Text Telescope: Text-Focused Scene Image Super-Resolution

• ECCV2020:PlugNet: Degradation Aware Scene Text Recognition Supervised by a Pluggable Super-Resolution Unit

  • 可插拔超分模组

---

Diffusion models

• arXiv-2023:GlyphDraw: Seamlessly Rendering Text with Intricate Spatial Structures in Text-to-Image Generation
• arXiv-2023:TextDiffuser: Diffusion Models as Text Painters
• arXiv-2023:DiffUTE: Universal Text Editing Diffusion Model
• arXiv-2023:GlyphControl: Glyph Conditional Control for Visual Text Generation

---

Others

• PR-2023:Towards open-set text recognition via label-to-prototype learning
• ECCV-2022:TextAdaIN: Paying Attention to Shortcut Learning in Text Recognizers
• TMM-2022:Dual Relation Network for Scene Text Recognition
• ECCV-2022:Background-Insensitive Scene Text Recognition with Text Semantic Segmentation
• arXiv-2022:A Scene-Text Synthesis Engine Achieved Through Learning from Decomposed Real-World Data
• arXiv-2022:Scene Text Recognition with Single-Point Decoding Network
• ECCV-2022:Toward Understanding WordArt: Corner-Guided Transformer for Scene Text Recognition
• arXiv-2022:Invariant Autoencoders for Text Recognition and Document Enhancement
• arXiv-2022:Training Protocol Matters: Towards Accurate Scene Text Recognition via Training Protocol Searching
  • 通过搜索训练参数来提升现有模型性能
• arXiv-2022:Text-DIAE: Degradation Invariant Autoencoders for Text Recognition and Document Enhancement
• arXiv-2022:Invariant Autoencoders for Text Recognition and Document Enhancement
• arXiv-2022:Towards Open-Set Text Recognition via Label-to-Prototype Learning
  • 当测试阶段遇到训练集中没有出现过的字符时，应该如何应对，场景文字识别中的开集问题
• AAAI-2022:FedOCR: Efficient and Secure Federated Learning for Scene Text Recognition
  • 联邦学习用于场景文字识别
• AAAI-2021,引用数::SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition
  • 图片送入网络前先在颜色上进行矫正
• arXiv-2021:Revisiting Classification Perspective on Scene Text Recognition
  • 把文本识别当作一个图像分类任务
• arXiv-2020:Hamming OCR: A Locality Sensitive Hashing Neural Network for Scene Text Recognition
  • 当识别种类数增大时，softmax embedding层就会更大，计算量也就会增大。本文提出使用汉明编码来进行解码，而不是使用one-hot进行解码
• ICCV Workshop-2021:Meta Self-Learning for Multi-Source Domain Adaptation: A Benchmark
  • 构建了一个多域的中文数据集，定义为Domain Adaptation问题
• ICCV-2021, 引用数:3:Towards the Unseen: Iterative Text Recognition by Distilling from Errors
  • 重复学习预测错误的样本
• CVPR-2021, 引用数:1:Primitive Representation Learning for Scene Text Recognition
  • 表征学习
• CVPR-2021, 引用数:1:What If We Only Use Real Datasets for Scene Text Recognition? Toward Scene Text Recognition With Fewer Labels
  • 如果只用真实数据训练识别网络会怎样？
• ACM MM-2020, 引用数:6:Exploring Font-independent Features for Scene Text Recognition
  • 考虑STR中字体风格的问题，用GAN将字体归一化进行识别
• CVPR-2020, 引用数:15:What Machines See Is Not What They Get: Fooling Scene Text Recognition Models with Adversarial Text Images
  • 探究STR中的对抗攻击问题
• CVPR-2020, 引用数:11:On Vocabulary Reliance in Scene Text Recognition
  • 在合成数据集上训练的识别器有字典依赖问题，本文探讨相关解决对策
• IJCV-2020, 引用数：6:Separating content from style using adversarial learning for recognizing text in the wild
  • 使用对抗生成网络把文字从背景中分离出来进行识别
• CVPR-2019，引用数:56:Sequence-to-sequence domain adaptation network for robust text image recognition
  • 探讨STR中的域自适应问题
• CVPR-2019, 引用数:49:Aggregation Cross-Entropy for Sequence Recognition
  • 提出了一种全新的聚合交叉熵损失，通过计数的方法做序列识别，速度很快

Conference

CVPR

• CVPR2022:Pushing the Performance Limit of Scene Text Recognizer without Human Annotation
• CVPR2022:Syntax-Aware Network for Handwritten Mathematical Expression Recognition
• CVPR2022:Open-set Text Recognition via Character-Context Decoupling
• CVPR2022:A Text Attention Network for Spatial Deformation Robust Scene Text Image Super-resolution
  • 场景文字超分
• CVPR2022:SimAN: Exploring Self-Supervised Representation Learning of Scene Text via Similarity-Aware Normalization
  • 生成式的对比学习
• CVPR2021:Scene Text Telescope: Text-Focused Scene Image Super-Resolution
• CVPR-2021, 引用数:5:Sequence-to-Sequence Contrastive Learning for Text Recognition
  • 首次在STR中引入对比学习的方法
• CVPR-2021, 引用数:1:Primitive Representation Learning for Scene Text Recognition
  • 表征学习
• CVPR-2021, 引用数:1:What If We Only Use Real Datasets for Scene Text Recognition? Toward Scene Text Recognition With Fewer Labels
  • 如果只用真实数据训练识别网络会怎样？
• CVPR-2021，Oral，引用数:1 Read Like Humans: Autonomous, Bidirectional and Iterative Language Modeling for Scene Text Recognition
  • ABINet
  • 屠榜作品，在SRN基础进行改良，从人类阅读的角度进行思考
• CVPR-2020, 引用数:42:SCATTER: selective context attentional scene text recognizer
  • 多阶段
• CVPR-2020, 引用数:58 Towards accurate scene text recognition with semantic reasoning networks
  • SRN
  • 引入Transformer进行语言建模，视觉+语言模型，取得SOTA效果
• CVPR-2020，引用数:59 Seed: Semantics enhanced encoder-decoder framework for scene text recognition
  • SEED
  • 在BiLSTM第一个单元前输入语义信息
  • 首次尝试把语言模型引入场景文字识别中
• CVPR-2020, 引用数:21:UnrealText: Synthesizing realistic scene text images from the unreal world
  • 虚幻引擎来渲染文字图像
• CVPR-2020，引用数:32 Learn to Augment: Joint Data Augmentation and Network Optimization for Text Recognition
• 利用仿射变化对图像进行数据增广，有效提升准确率
• CVPR-2020, 引用数:15:What Machines See Is Not What They Get: Fooling Scene Text Recognition Models with Adversarial Text Images
  • 探究STR中的对抗攻击问题
• CVPR-2020, 引用数:11:On Vocabulary Reliance in Scene Text Recognition
• CVPRWorkshop-2020, 引用数:28:On Recognizing Texts of Arbitrary Shapes with 2D Self-Attention
  • Transformer encoder + decoder, 引入自适应的2D位置编码，对于旋转文字，多行文字有较好的鲁棒性
  • 在合成数据集上训练的识别器有字典依赖问题，本文探讨相关解决对策
• CVPR-2019, 引用数:165:ESIR: End-to-end scene text recognition via iterative image rectification
  • 迭代矫正
• CVPR-2019，引用数:56:Sequence-to-sequence domain adaptation network for robust text image recognition
  • 探讨STR中的域自适应问题
• CVPR-2019, 引用数:49:Aggregation Cross-Entropy for Sequence Recognition
  • 提出了一种全新的聚合交叉熵损失，通过计数的方法做序列识别，速度很快
• CVPR-2018, 引用数:93:Edit probability for scene text recognition
  • 现有的attention方法采用最大似然损失函数，本文探讨输出概率分布和预测间的关系
• CVPR-2018,引用数:196:AON: Towards arbitrarily-oriented text recognition
  • 关注于不规则文字的识别
• CVPR-2016, 引用数:370:Recursive recurrent nets with attention modeling for OCR in the wild
• CVPR-2016, 引用数:415:Robust Scene Text Recognition With Automatic Rectification
  • TPS矫正+Attention
• CVPR-2016, 引用数:979:Synthetic data for text localisation in natural images
  • SynthText数据集

---

ICCV

• ICCV-2023: Self-supervised Character-to-Character Distillation for Text Recognition
• ICCV-2023: A Benchmark for Chinese-English Scene Text Image Super-resolution
• ICCV-2023: MRN: Multiplexed Routing Network for Incremental Multilingual Text Recognition
• ICCV-2023: Revisiting Scene Text Recognition: A Data Perspective
• ICCV Workshop-2021:Meta Self-Learning for Multi-Source Domain Adaptation: A Benchmark
  • 构建了一个多域的中文数据集，定义为Domain Adaptation问题
• ICCV-2021, 引用数:3:Towards the Unseen: Iterative Text Recognition by Distilling from Errors
  • 重复学习预测错误的样本
• ICCV-2021，引用数:1 From Two to One: A New Scene Text Recognizer with Visual Language Modeling Network
  • VisionLAN
  • 提出了一个新的遮挡文字数据集
  • 弱监督的将语言模型融入进视觉模型中
• ICCV-2021，引用数:1 Joint Visual Semantic Reasoning: Multi-Stage Decoder for Text Recognition
  • 多阶段+transformer识别器
  • 引入gumbel softmax，解决视觉到语义不可导问题
• ICCV-2019, 引用数:204:What is wrong with scene text recognition model comparisons? dataset and model analysis
  • 框架型文章，值得一读
• ICCV-2019, 引用数:77:Symmetry-constrained Rectification Network for Scene Text Recognition
  • 带限制的矫正网络
• ICCV-2017, 引用数:290:Focusing attention: Towards accurate text recognition in natural images
  • attention存在注意力飘移问题，引入Focusing Network把飘移的注意力抓回来

---

ECCV

• ECCV-2022:TextAdaIN: Paying Attention to Shortcut Learning in Text Recognizers
• ECCV-2022: Pure Transformer with Integrated Experts for Scene Text Recognition
• ECCV-2022:Background-Insensitive Scene Text Recognition with Text Semantic Segmentation
• ECCV-2022:Levenshtein OCR
• ECCV-2022:Multi-Granularity Prediction for Scene Text Recognition
• ECCV-2022:SGBANet: Semantic GAN and Balanced Attention Network for Arbitrarily Oriented Scene Text Recognition
• ECCV-2022:Scene Text Recognition with Permuted Autoregressive Sequence Models
• ECCV-2020, 引用数:27:Robustscanner: Dynamically enhancing positional clues for robust text recognition
• ECCV-2020,引用数:2:Variational Connectionist Temporal Classification
  • 提出变分CTC来增强网络对于非blank符号的学习
• ECCV-2018, 引用数:69:Synthetically supervised feature learning for scene text recognition

AAAI

• AAAI-2022: Perceiving Stroke-Semantic Context: Hierarchical Contrastive Learning for Robust Scene Text Recognition
• AAAI-2022:Visual Semantics Allow for Textual Reasoning Better in Scene Text Recognition
• AAAI-2022:Context-based Contrastive Learning for Scene Text Recognition
  • 对比学习用于场景文字识别
• AAAI-2022:FedOCR: Efficient and Secure Federated Learning for Scene Text Recognition
• AAAI-2022:Visual Semantics Allow for Textual Reasoning Better in Scene Text Recognition
  • 提升语言模型对于任意现状文本的识别能力
• AAAI-2022:Text Gestalt: Stroke-Aware Scene Text Image Super-Resolution
  • 场景文字超分，引入笔画级别的监督
• AAAI-2021,引用数::SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition
  • 图片送入网络前先在颜色上进行矫正
• AAAI-2020,引用数:27:GTC: Guided Training of CTC towards Efficient and Accurate Scene Text Recognition
  • 训练的时候用Attention分支辅助CTC，测试的时候只用CTC
• AAAI-2020, 引用数：28:Textscanner: Reading characters in order for robust scene text recognition
  • 实例分割
• AAAI-2020, 引用数：67:Decoupled attention network for text recognition
  • 考虑到注意力机制对于长文本的飘移，采用UNet架构来直接生成注意力图，将注意力图和解码之间解耦开
• AAAI-2019，引用数:128:Show, attend and read: A simple and strong baseline for irregular text recognition
  • 2D Attention
  • 在Attention计算过程中引入门控机制
• AAAI-2018, 引用数:103:Char-Net: A character-aware neural network for distorted scene text recognition
  • 加入单字检测分支。
• AAAI-2018, 引用数:50:SEE: Towards Semi-Supervised End-to-End Scene Text Recognition
  • 半监督端到端识别。

---

NIPS

• NIPS-2018,引用数:25:Connectionist Temporal Classification with Maximum Entropy Regularization
• NIPS-2017, 引用数:105:Gated recurrent convolution neural network for OCR
  • GRCNN
• NIPS-2014, 引用数:737:Synthetic data and artificial neural net?works for natural scene text recognition
  • MJ数据集

---

Others

• BMVC-2022:Visual-semantic transformer for scene text recognition
• BMVC-2022:Parallel and Robust Text Rectifier for Scene Text Recognition
• ICFHR-2022:A Vision Transformer Based Scene Text Recognizer with Multi-grained Encoding and Decoding
• ECCV-2022:Toward Understanding WordArt: Corner-Guided Transformer for Scene Text Recognition
• ECCV-2022:SGBANet: Semantic GAN and Balanced Attention Network for Arbitrarily Oriented Scene Text Recognition
  • Attention的解码时候的对齐是没有限制的，故引入CTC对Attention的对齐进行监督
• ACM MM-2018, 引用数:44:Attention and language ensemble for scene text recognition with convolutional sequence modeling
• IJCAI-2017, 引用数:124:Learning to read irregular text with attention mechanisms
• BMVC-2016,引用数:136:STAR-Net: A spatial attention residue network for scene text recognition

Journal

TPAMI

• TPAMI-2018, 引用数:329:ASTER: An Attentional Scene Text Recognizer with Flexible Rectification
  • Attention+Rectification经典之作
• TPAMI-2016,引用数:1497:An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
  • 场景文字识别开山之作，引入CTC将识别变为一个序列到序列的问题。

---

Pattern Recognition

• PR-2023:Towards open-set text recognition via label-to-prototype learning
• Patter Recognition-2021,引用数:23:Master: Multi-aspect non-local network for scene text recognition
• Pattern Recognition-2020,引用数:3:Reinterpreting CTC training as iterative fitting
  • 探讨CTC数学原理，将CTC Loss解释为交叉熵损失，较为理论
• Pattern Recognition-2018, 引用数:161MORAN: A Multi-Object Rectified Attention Network for Scene Text Recognition
  • 任意方向矫正，效果比ASTER出色
• Pattern Recognition-2017, 引用数:99:Accurate recognition of words in scenes without character segmentation using recurrent neural network

Others

• ICCV-2023: MRN: Multiplexed Routing Network for Incremental Multilingual Text Recognition
• IJCAI-2023:TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition
• IJCAI-2022:Linguistic More: Taking a Further Step toward Efficient and Accurate Scene Text Recognition
• ICDAR-2023:Scene Text Recognition with Image-Text Matching-guided Dictionary
• arXiv-2023:Improving Scene Text Recognition for Character-Level Long-Tailed Distribution
• TIP-2023:Text prior guided scene text image super-resolution
• Neurocomputing-2023:DPF-S2S: A novel dual-pathway-fusion-based sequence-to-sequence text recognition model
• WACV-2023:Seq-UPS: Sequential Uncertainty-aware Pseudo-label Selection for Semi-Supervised Text Recognition
• arXiv-2023:CLIPTER: Looking at the Bigger Picture in Scene Text Recognition
• ECCVW-2022:On calibration of scene-text recognition models
• Others:STR transformer: a cross-domain transformer for scene text recognition
• BMCV-2022: Masked Vision-Language Transformers for Scene Text Recognition
• Applied intelligence:Scene text recognition based on two-stage attention and multi-branch feature fusion module
• ICPR-2022: Portmanteauing Features for Scene Text Recognition
• ACCESS-2022:Scene Text Recognition with Semantics
• TIP-2022:PETR: Rethinking the Capability of Transformer-Based Language Model in Scene Text Recognition
• TMM-2022:Dual Relation Network for Scene Text Recognition
• IJCV-2021, 引用数:156:Scene text detection and recognition: The deep learning era
• Neural Computing-2020, 引用数:17:Adaptive embedding gate for attention-based scene text recognition
• IEEE Access-2019,引用数:30:Natural Scene Text Recognition Based on Encoder-Decoder Framework
• Neural Computing-2018, 引用数:41:Reading scene text with fully convolutional sequence modeling
  • Attention算法使用RNN建模，计算复杂并且较难训练，本文使用全卷积网络来捕获全局信息，比BiLSTM更加有效