Accurate Scene Text Detection through Border Semantics Awareness and Bootstrapping
This addresses accurate text localization in scenes, particularly for long words or lines, with incremental improvements over existing methods.
The paper tackles scene text detection by introducing bootstrapping to handle limited training data and improve consistency for long text, and semantics-aware border detection for more accurate localization. It achieves state-of-the-art results, including 80.1 F-score on MSRA-TD500 and 67.1 F-score on ICDAR2017-RCTW.
This paper presents a scene text detection technique that exploits bootstrapping and text border semantics for accurate localization of texts in scenes. A novel bootstrapping technique is designed which samples multiple 'subsections' of a word or text line and accordingly relieves the constraint of limited training data effectively. At the same time, the repeated sampling of text 'subsections' improves the consistency of the predicted text feature maps which is critical in predicting a single complete instead of multiple broken boxes for long words or text lines. In addition, a semantics-aware text border detection technique is designed which produces four types of text border segments for each scene text. With semantics-aware text borders, scene texts can be localized more accurately by regressing text pixels around the ends of words or text lines instead of all text pixels which often leads to inaccurate localization while dealing with long words or text lines. Extensive experiments demonstrate the effectiveness of the proposed techniques, and superior performance is obtained over several public datasets, e. g. 80.1 f-score for the MSRA-TD500, 67.1 f-score for the ICDAR2017-RCTW, etc.