Scene Text Detection for Augmented Reality – Character Bigram Approach to reduce False Positive Rate

12/26/2020
by   Sagar Gubbi, et al.
0

Natural scene text detection is an important aspect of scene understanding and could be a useful tool in building engaging augmented reality applications. In this work, we address the problem of false positives in text spotting. We propose improving the performace of sliding window text spotters by looking for character pairs (bigrams) rather than single characters. An efficient convolutional neural network is designed and trained to detect bigrams. The proposed detector reduces false positive rate by 28.16 dataset. We demonstrate that detecting bigrams is a computationally inexpensive way to improve sliding window text spotters.

READ FULL TEXT
research
01/12/2021

Using uncertainty estimation to reduce false positives in liver lesion detection

Despite the successes of deep learning techniques at detecting objects i...
research
11/21/2022

Rooms with Text: A Dataset for Overlaying Text Detection

In this paper, we introduce a new dataset of room interior pictures with...
research
11/02/2021

Towards Text-based Phishing Detection

This paper reports on an experiment into text-based phishing detection u...
research
11/12/2017

Feature Enhancement Network: A Refined Scene Text Detector

In this paper, we propose a refined scene text detector with a novel Fea...
research
01/02/2019

Extension of the Geometric Mean Constant False Alarm Rate Detector to Multiple Pulses

The development of sliding window detection processes, based upon a sing...
research
12/23/2018

A Note on the Bayesian Approach to Sliding Window Detector Development

Recently a Bayesian methodology has been introduced, enabling the constr...
research
01/04/2019

Compensating for Interference in Sliding Window Detection Processes using a Bayesian Paradigm

Sliding window detectors are non-coherent decision processes, designed i...

Please sign up or login with your details

Forgot password? Click here to reset