Accurate Scene Text Detection through Border Semantics Awareness and Bootstrapping
Accurate Scene Text Detection through Border Semantics Awareness and Bootstrapping
复制标题
DOI：
10.1007/978-3-030-01270-0_22
复制
发表时间：
2018-07
期刊：
影响因子：
--
通讯作者：
Chuhui Xue;Shijian Lu;Fangneng Zhan
中科院分区：
文献类型：
--
作者：
Chuhui Xue;Shijian Lu;Fangneng Zhan
本文提出了一种场景文本检测技术，利用自举和文本边界语义准确定位的文本场景。设计了一种新的自举技术，该技术对一个单词或文本行的多个“子部分”进行采样，从而有效地缓解了有限训练数据的约束。同时，文本“子部分”的重复采样提高了预测文本特征图的一致性，这对于预测长单词或文本行的单个完整框而不是多个破框至关重要。此外，设计了一种语义感知的文本边界检测技术，该技术为每个场景文本产生四种类型的文本边界段。使用语义感知的文本边界，场景文本可以通过在单词或文本行的结尾周围回归文本像素而不是所有文本像素来更准确地定位，这在处理长单词或文本行时通常会导致不准确的定位。大量的实验证明了所提出的技术的有效性，并且在几个公共数据集上获得了上级性能，例如MSRA-TD 500的f分数为80.1，ICDAR 2017-RCTW的f分数为67.1等。
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, eg 80.1 f-score for the MSRA-TD500, 67.1 f-score for the ICDAR2017-RCTW, etc.