WordSup: Exploiting Word Annotations for Character based Text Detection

Abstract

Imagery texts are usually organized as a hierarchy of several visual elements, i.e. characters, words, text lines and text blocks. Among these elements, character is the most basic one for various languages such as Western, Chinese, Japanese, mathematical expression and etc. It is natural and convenient to construct a common text detection engine based on character detectors. However, training character detectors requires a vast of location annotated characters, which are expensive to obtain. Actually, the existing real text datasets are mostly annotated in word or line level. To remedy this dilemma, we propose a weakly supervised framework that can utilize word annotations, either in tight quadrangles or the more loose bounding boxes, for character detector training. When applied in scene text detection, we are thus able to train a robust character detector by exploiting word annotations in the rich large-scale real scene text datasets, e.g. ICDAR15 and COCO-text. The character detector acts as a key role in the pipeline of our text detection engine. It achieves the state-of-the-art performance on several challenging scene text detection benchmarks. We also demonstrate the flexibility of our pipeline by various scenarios, including deformed text detection and math expression recognition.

2017 International Conference on Computer Vision

References in corpus (17)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Fully Convolutional Networks for Semantic Segmentation
• Synthetic Data and Artificial Neural Networks for Natural Scene Text Recognition
• Robust Text Detection in Natural Scene Images
• Feature Pyramid Networks for Object Detection
• DenseBox: Unifying Landmark Localization with End to End Object Detection
• Text-Attentional Convolutional Neural Networks for Scene Text Detection
• COCO-Text: Dataset and Benchmark for Text Detection and Recognition in Natural Images
• Scene Text Detection via Holistic, Multi-Channel Prediction
• Synthetic Data for Text Localisation in Natural Images
• BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation
• HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection
• EAST: An Efficient and Accurate Scene Text Detector
• 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
• Detecting Oriented Text in Natural Images by Linking Segments

Cited by in corpus (17)

• A Single-Shot Arbitrarily-Shaped Text Detector based on Context Attended Multi-Task Learning
• Selective Feature Connection Mechanism: Concatenating Multi-layer CNN Features with a Feature Selector
• Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation
• Sliding Line Point Regression for Shape Robust Scene Text Detection
• A pooling based scene text proposal technique for scene text reading in the wild
• Scene Text Detection with Supervised Pyramid Context Network
• Rotation-Sensitive Regression for Oriented Scene Text Detection
• An Anchor-Free Region Proposal Network for Faster R-CNN based Text Detection Approaches
• MSR: Multi-Scale Shape Regression for Scene Text Detection
• Geometry Normalization Networks for Accurate Scene Text Detection
• FOTS: Fast Oriented Text Spotting with a Unified Network
• You Only Recognize Once: Towards Fast Video Text Spotting
• DeRPN: Taking a further step toward more general object detection
• TextNet: Irregular Text Reading from Images with an End-to-End Trainable Network
• Detecting Text in the Wild with Deep Character Embedding Network
• Correlation Propagation Networks for Scene Text Detection
• Accurate Scene Text Detection through Border Semantics Awareness and Bootstrapping