Rethinking Irregular Scene Text Recognition
This work addresses the challenge of reading text from natural images with distortions for computer vision applications, but it is incremental as it builds on existing rectification methods.
The paper tackles the problem of irregular scene text recognition by improving rectification-based methods with a bag of tricks, achieving state-of-the-art accuracies of 89.6% on CUTE-80 and 76.3% on Total-Text, with improvements of 6.3% and 14.7% respectively, and winning the ICDAR 2019 Arbitrary-Shaped Text Challenge with 74.3% accuracy.
Reading text from natural images is challenging due to the great variety in text font, color, size, complex background and etc.. The perspective distortion and non-linear spatial arrangement of characters make it further difficult. While rectification based method is intuitively grounded and has pushed the envelope by far, its potential is far from being well exploited. In this paper, we present a bag of tricks that prove to significantly improve the performance of rectification based method. On curved text dataset, our method achieves an accuracy of 89.6% on CUTE-80 and 76.3% on Total-Text, an improvement over previous state-of-the-art by 6.3% and 14.7% respectively. Furthermore, our combination of tricks helps us win the ICDAR 2019 Arbitrary-Shaped Text Challenge (Latin script), achieving an accuracy of 74.3% on the held-out test set. We release our code as well as data samples for further exploration at https://github.com/Jyouhou/ICDAR2019-ArT-Recognition-Alchemy