Accurate Scene Text Detection through Border Semantics Awareness and Bootstrapping

Abstract

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.

arXiv comment14 pages, 8 figures, accepted by ECCV 2018

References in corpus (18)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Random Erasing Data Augmentation
• R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection
• DenseBox: Unifying Landmark Localization with End to End Object Detection
• TextBoxes: A Fast Text Detector with a Single Deep Neural Network
• Text-Attentional Convolutional Neural Networks for Scene Text Detection
• Scene Text Detection via Holistic, Multi-Channel Prediction
• Synthetic Data for Text Localisation in Natural Images
• Multi-Oriented Text Detection with Fully Convolutional Networks
• Accurate Text Localization in Natural Image with Cascaded Convolutional Text Network
• Deep Matching Prior Network: Toward Tighter Multi-oriented Text Detection
• Single Shot Text Detector with Regional Attention
• WordSup: Exploiting Word Annotations for Character based Text Detection
• An end-to-end TextSpotter with Explicit Alignment and Attention
• ICDAR2017 Competition on Reading Chinese Text in the Wild (RCTW-17)
• WeText: Scene Text Detection under Weak Supervision
• WordFence: Text Detection in Natural Images with Border Awareness