CASIA OpenIR  > 语音交互团队
Deep imitator: handwriting calligraphy imitation via deep attention networks
Zhao BC(赵博程); Tao JH(陶建华); Yang MH(杨明浩); Tian ZK(田正坤); Fan CH(范存航); Bai Y(白烨)
Source PublicationPattern Recogniton
2019-10-12
Issue已接收Pages:已接收
Abstract

Calligraphy imitation (CI) from a handful of target handwriting samples is
such a challenging task that most of the existing writing style analysis or
handwriting generation methods do not exhibit satisfactory performance. In this paper, we propose a novel multi-module framework to address the problem of CI. Firstly, we utilized a deep convolution neural network (CNN) to extract personalized calligraphical features. Then we built a calligraphyclustering attention module and a mata-style matrix (msM) to compute an
embedding of calligraphy. The structure of conditional gated recurrent unit (cGRU) is then improved to predict the probabilistic density of pen tip movement displacement by dual condition inputs. Finally, we generated personalized handwriting stroke sequences through iterative sampling with Gaussian mixture model (GMM). Experiments on public online handwriting databases verify that the proposed method could achieve satisfactory performance; the generated samples achieved high similarities with original handwriting examples.

Keywordcalligraphy imitation, attention, mata-style matrix, condition Gated Recurrent Unit
Indexed BySCI
Language英语
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/28352
Collection语音交互团队
Recommended Citation
GB/T 7714
Zhao BC,Tao JH,Yang MH,et al. Deep imitator: handwriting calligraphy imitation via deep attention networks[J]. Pattern Recogniton,2019(已接收):已接收.
APA Zhao BC,Tao JH,Yang MH,Tian ZK,Fan CH,&Bai Y.(2019).Deep imitator: handwriting calligraphy imitation via deep attention networks.Pattern Recogniton(已接收),已接收.
MLA Zhao BC,et al."Deep imitator: handwriting calligraphy imitation via deep attention networks".Pattern Recogniton .已接收(2019):已接收.
Files in This Item: Download All
File Name/Size DocType Version Access License
elsarticle-template-(2498KB)期刊 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhao BC(赵博程)]'s Articles
[Tao JH(陶建华)]'s Articles
[Yang MH(杨明浩)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhao BC(赵博程)]'s Articles
[Tao JH(陶建华)]'s Articles
[Yang MH(杨明浩)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhao BC(赵博程)]'s Articles
[Tao JH(陶建华)]'s Articles
[Yang MH(杨明浩)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: elsarticle-template-num-updateversion.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.