Scene Text Recognition2020en
icdar-2003
Dataset from Papers With Code
Metrics:accuracy, cer, wer, f1
Current State of the Art
Yet Another Text Recognizer
Unknown
97.1
accuracy
accuracy Progress Over Time
Showing 6 breakthroughs from Jul 2015 to Jul 2021
Key Milestones
Jul 2015
CRNN
From paper: An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition
89.4
Apr 2019
Baek et al.
From paper: What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis
94.4
+3.2%
Oct 2019
SATRN
From paper: On Recognizing Texts of Arbitrary Shapes with 2D Self-Attention
96.7
+2.4%
Jul 2021
Yet Another Text RecognizerCurrent SOTA
From paper: Why You Should Try the Real Data for the Scene Text Recognition
97.1
+0.4%
Total Improvement
8.6%
Time Span
6y 1m
Breakthroughs
6
Current SOTA
97.1
Top Models Performance Comparison
Top 10 models ranked by accuracy
Best Score
97.1
Top Model
Yet Another Text ...
Models Compared
10
Score Range
7.0
accuracyPrimary
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | Yet Another Text Recognizer | 97.1 | Jul 2021 | |
| 2 | SIGA_T | 97 | Mar 2022 | |
| 3 | SATRN | 96.7 | Oct 2019 | |
| 4 | DAN | 95 | Dec 2019 | |
| 5 | SAFL | 95 | Jan 2022 | |
| 6 | CSTR | 94.8 | Feb 2021 | |
| 7 | Baek et al. | 94.4 | Apr 2019 | |
| 8 | ViTSTR | 94.3 | May 2021 | |
| 9 | AON | 91.5 | Nov 2017 | |
| 10 | RARE | 90.1 | Mar 2016 | |
| 11 | STAR-Net | 89.9 | Star-net: A spatial attention residue network for scene text recognition.Code | Sep 2016 |
| 12 | CRNN | 89.4 | Jul 2015 |
Related Papers11
Self-supervised Implicit Glyph Attention for Text Recognition
Mar 2022Models: SIGA_T
SAFL: A Self-Attention Scene Text Recognizer with Focal Loss
Jan 2022Models: SAFL
Why You Should Try the Real Data for the Scene Text Recognition
Jul 2021Models: Yet Another Text Recognizer
Vision Transformer for Fast and Efficient Scene Text Recognition
May 2021Models: ViTSTR
Revisiting Classification Perspective on Scene Text Recognition
Feb 2021Models: CSTR
Decoupled Attention Network for Text Recognition
Dec 2019Models: DAN
On Recognizing Texts of Arbitrary Shapes with 2D Self-Attention
Oct 2019Models: SATRN
What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis
Apr 2019Models: Baek et al.
AON: Towards Arbitrarily-Oriented Text Recognition
Nov 2017Models: AON
Robust Scene Text Recognition with Automatic Rectification
Mar 2016Models: RARE