A Multi-oriented Chinese Keyword Spotter Guided by Text Line Detection

TL;DR

This paper introduces a novel Chinese keyword spotting method in natural images that leverages text line detection and segmentation, inspired by Mask R-CNN, to improve accuracy in the absence of visual word boundaries.

Contribution

The paper presents a new Chinese keyword spotter that predicts keyword masks guided by text line detection, combining text line proposals with segmentation for improved performance.

Findings

Effective keyword spotting demonstrated on RCTW-17 and ICPR MTWI2018 datasets.

Parallel prediction of text lines and keywords enhances detection accuracy.

Method outperforms existing approaches in Chinese keyword spotting.

Abstract

Chinese keyword spotting is a challenging task as there is no visual blank for Chinese words. Different from English words which are split naturally by visual blanks, Chinese words are generally split only by semantic information. In this paper, we propose a new Chinese keyword spotter for natural images, which is inspired by Mask R-CNN. We propose to predict the keyword masks guided by text line detection. Firstly, proposals of text lines are generated by Faster R-CNN;Then, text line masks and keyword masks are predicted by segmentation in the proposals. In this way, the text lines and keywords are predicted in parallel. We create two Chinese keyword datasets based on RCTW-17 and ICPR MTWI2018 to verify the effectiveness of our method.