Why You Should Try the Real Data for the Scene Text Recognition

29 Jul 2021  ·  Vladimir Loginov ·

Recent works in the text recognition area have pushed forward the recognition results to the new horizons. But for a long time a lack of large human-labeled natural text recognition datasets has been forcing researchers to use synthetic data for training text recognition models. Even though synthetic datasets are very large (MJSynth and SynthTest, two most famous synthetic datasets, have several million images each), their diversity could be insufficient, compared to natural datasets like ICDAR and others. Fortunately, the recently released text-recognition annotation for OpenImages V5 dataset has comparable with synthetic dataset number of instances and more diverse examples. We have used this annotation with a Text Recognition head architecture from the Yet Another Mask Text Spotter and got comparable to the SOTA results. On some datasets we have even outperformed previous SOTA models. In this paper we also introduce a text recognition model. The model's code is available.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Scene Text Recognition ICDAR 2003 Yet Another Text Recognizer Accuracy 97.1 # 1
Scene Text Recognition ICDAR2013 Yet Another Text Recognizer Accuracy 96.8 # 12
Scene Text Recognition ICDAR2015 Yet Another Text Recognizer Accuracy 80.2 # 10
Scene Text Recognition SVT Yet Another Text Recognizer Accuracy 94.7 # 9


No methods listed for this paper. Add relevant methods here