This paper reports the ICDAR2019 Robust Reading Challenge on Arbitrary-Shaped Text (RRC-ArT) that consists of three major challenges: i) scene text detection, ii) scene text recognition, and iii) scene text spotting.
Deep learning based methods have achieved surprising progress in Scene Text Recognition (STR), one of classic problems in computer vision.
Incidental scene text spotting is considered one of the most difficult and valuable challenges in the document analysis community.
#2 best model for Scene Text Detection on IC15
Recently, models based on deep neural networks have dominated the fields of scene text detection and recognition.
#2 best model for Scene Text Detection on ICDAR 2013
We present a scenario where semantic similarity is not enough, and we devise a neural approach to learn semantic relatedness.
We propose a post-processing approach to improve scene text recognition accuracy by using occurrence probabilities of words (unigram language model), and the semantic correlation between scene and text.