One-Shot Chinese Character Recognition Based on Deep Siamese Networks

Huichao Li, Xuemei Ren*, Yongfeng Lv

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Citations (Scopus)

Abstract

This paper applies deep siamese network to one-shot Chinese handwritten character recognition. Different from common image classification tasks, the CASIA HWDB1.1 dataset used here contains more than 3000 categories, with only few samples in each one. We propose a basic deep siamese model as well as an improved model with multi-layer features mechanism and batch normalization for extracting the similarity of the input pairs, and implement one-shot recognition by categorizing the test example to the class where the support sample is the most similar. Experiments prove that our model is able to recognize Chinese characters of unseen classes in training with only one support example efficiently.

Original languageEnglish
Title of host publicationProceedings of 2019 Chinese Intelligent Systems Conference - Volume I
EditorsYingmin Jia, Junping Du, Weicun Zhang
PublisherSpringer Verlag
Pages742-750
Number of pages9
ISBN (Print)9789813296817
DOIs
Publication statusPublished - 2020
EventChinese Intelligent Systems Conference, CISC 2019 - Haikou, China
Duration: 26 Oct 201927 Oct 2019

Publication series

NameLecture Notes in Electrical Engineering
Volume592
ISSN (Print)1876-1100
ISSN (Electronic)1876-1119

Conference

ConferenceChinese Intelligent Systems Conference, CISC 2019
Country/TerritoryChina
CityHaikou
Period26/10/1927/10/19

Keywords

  • CASIA dataset
  • Convolutional neural network
  • One-shot learning
  • Siamese network

Fingerprint

Dive into the research topics of 'One-Shot Chinese Character Recognition Based on Deep Siamese Networks'. Together they form a unique fingerprint.

Cite this