Deep Residual Text Detection Network for Scene Text

TL;DR

This paper introduces a deep residual network-based scene text detection method that leverages multi-level features and a vertical proposal mechanism, achieving state-of-the-art accuracy on ICDAR2013.

Contribution

The novel integration of ResNet with multi-level features and a vertical proposal mechanism enhances scene text detection performance.

Findings

Achieved F-measure of 0.91 on ICDAR2013

Outperforms previous state-of-the-art methods

Effective use of hierarchical convolutional features

Abstract

Scene text detection is a challenging problem in computer vision. In this paper, we propose a novel text detection network based on prevalent object detection frameworks. In order to obtain stronger semantic feature, we adopt ResNet as feature extraction layers and exploit multi-level feature by combining hierarchical convolutional networks. A vertical proposal mechanism is utilized to avoid proposal classification, while regression layer remains working to improve localization accuracy. Our approach evaluated on ICDAR2013 dataset achieves F-measure of 0.91, which outperforms previous state-of-the-art results in scene text detection.