Single Shot Text Detector with Regional Attention

Abstract

We present a novel single-shot text detector that directly outputs word-level bounding boxes in a natural image. We propose an attention mechanism which roughly identifies text regions via an automatically learned attentional map. This substantially suppresses background interference in the convolutional features, which is the key to producing accurate inference of words, particularly at extremely small sizes. This results in a single model that essentially works in a coarse-to-fine manner. It departs from recent FCN- based text detectors which cascade multiple FCN models to achieve an accurate prediction. Furthermore, we develop a hierarchical inception module which efficiently aggregates multi-scale inception features. This enhances local details, and also encodes strong context information, allow- ing the detector to work reliably on multi-scale and multi- orientation text with single-scale images. Our text detector achieves an F-measure of 77% on the ICDAR 2015 bench- mark, advancing the state-of-the-art results in [18, 28]. Demo is available at: http://sstd.whuang.org/.

To appear in IEEE International Conference on Computer Vision (ICCV), 2017

References in corpus (10)

• Caffe: Convolutional Architecture for Fast Feature Embedding
• Robust Text Detection in Natural Scene Images
• TextBoxes: A Fast Text Detector with a Single Deep Neural Network
• 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
• 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

Cited by in corpus (29)

• Attention Mechanisms in Computer Vision: A Survey
• Optimization for Arbitrary-Oriented Object Detection via Representation Invariance Loss
• Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network
• Weak Supervision for Generating Pixel-Level Annotations in Scene Text Segmentation
• Selective Feature Connection Mechanism: Concatenating Multi-layer CNN Features with a Feature Selector
• Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation
• Beyond Visual Semantics: Exploring the Role of Scene Text in Image Understanding
• Scene Text Synthesis for Efficient and Effective Deep Network Training
• Pixel-Anchor: A Fast Oriented Scene Text Detector with Combined Networks
• Sliding Line Point Regression for Shape Robust Scene Text Detection
• Scene Text Detection with Supervised Pyramid Context Network
• Rotation-Sensitive Regression for Oriented Scene Text Detection
• Feature Enhancement Network: A Refined Scene Text Detector
• MSR: Multi-Scale Shape Regression for Scene Text Detection
• TRIE: End-to-End Text Reading and Information Extraction for Document Understanding
• Verisimilar Image Synthesis for Accurate Detection and Recognition of Texts in Scenes
• Geometry Normalization Networks for Accurate Scene Text Detection
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• FOTS: Fast Oriented Text Spotting with a Unified Network
• GA-DAN: Geometry-Aware Domain Adaptation Network for Scene Text Detection and Recognition
• DeRPN: Taking a further step toward more general object detection
• Mask is All You Need: Rethinking Mask R-CNN for Dense and Arbitrary-Shaped Scene Text Detection
• Chargrid-OCR: End-to-end Trainable Optical Character Recognition for Printed Documents using Instance Segmentation
• Inductive Visual Localisation: Factorised Training for Superior Generalisation
• STELA: A Real-Time Scene Text Detector with Learned Anchor
• Accurate Scene Text Detection through Border Semantics Awareness and Bootstrapping
• Learning to Predict More Accurate Text Instances for Scene Text Detection
• Scene Text recognition with Full Normalization
• TextContourNet: a Flexible and Effective Framework for Improving Scene Text Detection Architecture with a Multi-task Cascade