Detecting Oriented Text in Natural Images by Linking Segments (2017-03-19T00:00:00.000000Z)

TL;DR

SegLink, an oriented text detection method to decompose text into two locally detectable elements, namely segments and links, achieves an f-measure of 75.0% on the standard ICDAR 2015 Incidental (Challenge 4) benchmark, outperforming the previous best by a large margin.

Abstract

Most state-of-the-art text detection methods are specific to horizontal Latin text and are not fast enough for real-time applications. We introduce Segment Linking (SegLink), an oriented text detection method. The main idea is to decompose text into two locally detectable elements, namely segments and links. A segment is an oriented box covering a part of a word or text line, A link connects two adjacent segments, indicating that they belong to the same word or text line. Both elements are detected densely at multiple scales by an end-to-end trained, fully-convolutional neural network. Final detections are produced by combining segments connected by links. Compared with previous methods, SegLink improves along the dimensions of accuracy, speed, and ease of training. It achieves an f-measure of 75.0% on the standard ICDAR 2015 Incidental (Challenge 4) benchmark, outperforming the previous best by a large margin. It runs at over 20 FPS on 512x512 images. Moreover, without modification, SegLink is able to detect long lines of non-Latin text, such as Chinese.

Authors

Serge J. Belongie

28 papers

X. Bai

25 papers

Baoguang Shi

4 papers

TL;DR

Abstract

Authors

References32 items

Robust Text Detection in Natural Scene Images

Detecting Text in Natural Image with Connectionist Text Proposal Network

Real-Time Lexicon-Free Scene Text Localization and Recognition

Scene Text Detection via Holistic, Multi-Channel Prediction

Synthetic Data for Text Localisation in Natural Images

Multi-oriented Text Detection with Fully Convolutional Networks

Training Region-Based Object Detectors with Online Hard Example Mining

SSD: Single Shot MultiBox Detector

FASText: Efficient Unconstrained Scene Text Detector

Multi-Orientation Scene Text Detection with Adaptive Clustering

ICDAR 2015 competition on Robust Reading

You Only Look Once: Unified, Real-Time Object Detection

Symmetry-based text line detection in natural scenes

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

Fast R-CNN

Efficient Scene text localization and recognition with local character refinement

Reading Text in the Wild with Convolutional Neural Networks

Fully convolutional networks for semantic segmentation

Deep Features for Text Spotting

Robust Scene Text Detection with Convolution Neural Network Induced MSER Trees

Very Deep Convolutional Networks for Large-Scale Image Recognition

Orientation Robust Text Line Detection in Natural Images

PhotoOCR: Reading Text in Uncontrolled Conditions

Text Localization in Natural Images Using Stroke Feature Transform and Text Covariance Descriptors

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation

ICDAR 2013 Robust Reading Competition

End-to-end text recognition with convolutional neural networks

Detecting texts of arbitrary orientations in natural images

Word Spotting in the Wild

Software available from tensorflow.org

Tensor-Flow: Large-scale machine learning on heterogeneous systems

Almazán, and L. de las Heras. ICDAR 2013 robust reading competition. In ICDAR 2013

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names