Deep Matching Prior Network: Toward Tighter Multi-oriented Text Detection

TL;DR

This paper introduces DMPNet, a CNN-based method for detecting multi-oriented scene text with tighter quadrilateral bounding boxes, improving accuracy over previous rectangle-based approaches.

Contribution

The paper proposes a novel CNN architecture with quadrilateral sliding windows, a Monte-Carlo based area computation, and a sequential regression protocol for precise multi-oriented text detection.

Findings

Achieved 70.64% F-measure on ICDAR 2015 dataset

Outperformed previous state-of-the-art with 63.76% F-measure

Demonstrated robustness and stability with the proposed smooth Ln loss

Abstract

Detecting incidental scene text is a challenging task because of multi-orientation, perspective distortion, and variation of text size, color and scale. Retrospective research has only focused on using rectangular bounding box or horizontal sliding window to localize text, which may result in redundant background noise, unnecessary overlap or even information loss. To address these issues, we propose a new Convolutional Neural Networks (CNNs) based method, named Deep Matching Prior Network (DMPNet), to detect text with tighter quadrangle. First, we use quadrilateral sliding windows in several specific intermediate convolutional layers to roughly recall the text with higher overlapping area and then a shared Monte-Carlo method is proposed for fast and accurate computing of the polygonal areas. After that, we designed a sequential protocol for relative regression which can exactly predict text with compact quadrangle. Moreover, a auxiliary smooth Ln loss is also proposed for further regressing the position of text, which has better overall performance than L2 loss and smooth L1 loss in terms of robustness and stability. The effectiveness of our approach is evaluated on a public word-level, multi-oriented scene text database, ICDAR 2015 Robust Reading Competition Challenge 4 "Incidental scene text localization". The performance of our method is evaluated by using F-measure and found to be 70.64%, outperforming the existing state-of-the-art method with F-measure 63.76%.