R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection

Abstract

In this paper, we propose a novel method called Rotational Region CNN (R2CNN) for detecting arbitrary-oriented texts in natural scene images. The framework is based on Faster R-CNN [1] architecture. First, we use the Region Proposal Network (RPN) to generate axis-aligned bounding boxes that enclose the texts with different orientations. Second, for each axis-aligned text box proposed by RPN, we extract its pooled features with different pooled sizes and the concatenated features are used to simultaneously predict the text/non-text score, axis-aligned box and inclined minimum area box. At last, we use an inclined non-maximum suppression to get the detection results. Our approach achieves competitive results on text detection benchmarks: ICDAR 2015 and ICDAR 2013.

8 pages, 6 figures, 3 tables

References in corpus (4)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• TextBoxes: A Fast Text Detector with a Single Deep Neural Network
• Deep Matching Prior Network: Toward Tighter Multi-oriented Text Detection
• Deep Direct Regression for Multi-Oriented Scene Text Detection

Cited by in corpus (32)

• LO-Det: Lightweight Oriented Object Detection in Remote Sensing Images
• Learning Modulated Loss for Rotated Object Detection
• Character Region Awareness for Text Detection
• Pyramid Mask Text Detector
• Improving the Detection of Small Oriented Objects in Aerial Images
• Sliding Line Point Regression for Shape Robust Scene Text Detection
• Text Detection and Recognition in the Wild: A Review
• Dynamic Anchor Learning for Arbitrary-Oriented Object Detection
• Multi-Grained Angle Representation for Remote Sensing Object Detection
• Quantifying Legibility of Indoor Spaces Using Deep Convolutional Neural Networks: Case Studies in Train Stations
• PIoU Loss: Towards Accurate Oriented Object Detection in Complex Environments
• Arbitrary-Shaped Text Detection withAdaptive Text Region Representation
• Dynamic Refinement Network for Oriented and Densely Packed Object Detection
• A Survey on Deep Domain Adaptation and Tiny Object Detection Challenges, Techniques and Datasets
• Convolutional Character Networks
• Oriented Object Detection in Aerial Images with Box Boundary-Aware Vectors
• FOTS: Fast Oriented Text Spotting with a Unified Network
• Sketch2code: Generating a website from a paper mockup
• Oriented R-CNN for Object Detection
• RRPN++: Guidance Towards More Accurate Scene Text Detection
• TextNet: Irregular Text Reading from Images with an End-to-End Trainable Network
• Learning Orientation-Estimation Convolutional Neural Network for Building Detection in Optical Remote Sensing Image
• Tracking Based Semi-Automatic Annotation for Scene Text Videos
• Object Detection based on OcSaFPN in Aerial Images with Noise
• TextTubes for Detecting Curved Text in the Wild
• Probabilistic Oriented Object Detection in Automotive Radar
• Towards Spatio-Temporal Video Scene Text Detection via Temporal Clustering
• Location-Aware Feature Selection Text Detection Network
• Scene Text Detection with Selected Anchor
• STELA: A Real-Time Scene Text Detector with Learned Anchor
• A Multiplexed Network for End-to-End, Multilingual OCR
• DGST : Discriminator Guided Scene Text detector