Character Region Attention For Text Spotting

A scene text spotter is composed of text detection and recognition modules. Many studies have been conducted to unify these modules into an end-to-end trainable model to achieve better performance. A typical architecture places detection and recognition modules into separate branches, and a RoI pool...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Baek, Youngmin, Shin, Seung, Baek, Jeonghun, Park, Sungrae, Lee, Junyeop, Nam, Daehyun, Lee, Hwalsuk
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 19.07.2020
Subjects
Architecture
Localization
Modules
Spatial data
Online AccessGet full text

Cover

More Information
Summary:A scene text spotter is composed of text detection and recognition modules. Many studies have been conducted to unify these modules into an end-to-end trainable model to achieve better performance. A typical architecture places detection and recognition modules into separate branches, and a RoI pooling is commonly used to let the branches share a visual feature. However, there still exists a chance of establishing a more complimentary connection between the modules when adopting recognizer that uses attention-based decoder and detector that represents spatial information of the character regions. This is possible since the two modules share a common sub-task which is to find the location of the character regions. Based on the insight, we construct a tightly coupled single pipeline model. This architecture is formed by utilizing detection outputs in the recognizer and propagating the recognition loss through the detection stage. The use of character score map helps the recognizer attend better to the character center points, and the recognition loss propagation to the detector module enhances the localization of the character regions. Also, a strengthened sharing stage allows feature rectification and boundary localization of arbitrary-shaped text regions. Extensive experiments demonstrate state-of-the-art performance in publicly available straight and curved benchmark dataset.
ISSN:2331-8422