Handwritten Chinese character recognition with spatial transformer and deep residual networks

This paper considers using deep neural networks for handwritten Chinese character recognition (HCCR) with arbitrary position, scale, and orientations. To solve this problem, we combine the recently proposed spatial transformer network (STN) with the deep residual network (DRN). The STN acts like a c...

Full description

Saved in:
Bibliographic Details
Published in2016 23rd International Conference on Pattern Recognition (ICPR) pp. 3440 - 3445
Main Authors Zhao Zhong, Xu-Yao Zhang, Fei Yin, Cheng-Lin Liu
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2016
Subjects
Character recognition
Data models
Generators
Neural networks
Robustness
Shape
Transforms
Online AccessGet full text
DOI10.1109/ICPR.2016.7900166

Cover

More Information
Summary:This paper considers using deep neural networks for handwritten Chinese character recognition (HCCR) with arbitrary position, scale, and orientations. To solve this problem, we combine the recently proposed spatial transformer network (STN) with the deep residual network (DRN). The STN acts like a character shape normalization procedure. Different from the traditional heuristic shape normalization methods, STN is learned directly from the data. Furthermore, the DRN makes the training of very deep network to be both efficient and effective. With the combination of STN and DRN, the whole model can be trained jointly in an end-to-end manner. In this paper, new state-of-the-art performance has been achieved by our proposed model on the offline ICDAR-2013 Chinese handwriting competition database. Moreover, the experiment on randomly distorted samples shows that the STN is very effective for robust HCCR in rectifying the shape of distorted characters.
DOI:10.1109/ICPR.2016.7900166