Immersive Traditional Chinese Portrait Painting: Research on Style Transfer and Face Replacement

Jiayue Li; Qing Wang; Shiji Li; Qiang Zhong; Qian Zhou

doi:10.1007/978-3-030-88007-1_16

Immersive Traditional Chinese Portrait Painting: Research on Style Transfer and Face Replacement

Jiayue Li, Qing Wang, Shiji Li, Qiang Zhong, Qian Zhou

China Agricultural University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

5 Scopus citations

Abstract

Traditional Chinese portrait is popular all over the world because of its unique oriental charm. However, how to use neural network to express the aesthetic and feelings in instantiated Chinese portrait effectively is still a challenging problem. This paper proposes a Photo to Chinese Portrait method (P-CP) providing immersive traditional Chinese portrait painting experience. Our method can produce two groups intriguing pictures. One is Chinese Portrait Style Picture, the other is Immersive Chinese Portrait Picture. We pay attention to neural style transfer for traditional Chinese portrait for the first time, and have trained a fast feedforward generative network to extract the corresponding style. The generative network principle is explored to guide the style transfer adjustment in detail. Face replacement is added to form a more appealing stylized effect. We also solve the problems of color and light violation, and unnatural seam. We hope this work offers a deeper and immersive conversation between modern society and antiques, and provides a useful step towards related interdisciplinary areas.

Original language	English
Title of host publication	Pattern Recognition and Computer Vision - 4th Chinese Conference, PRCV 2021, Proceedings
Editors	Huimin Ma, Liang Wang, Changshui Zhang, Fei Wu, Tieniu Tan, Yaonan Wang, Jianhuang Lai, Yao Zhao
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	192-203
Number of pages	12
ISBN (Print)	9783030880064
DOIs	https://doi.org/10.1007/978-3-030-88007-1_16
State	Published - 2021
Externally published	Yes
Event	4th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2021 - Beijing, China Duration: 29 Oct 2021 → 1 Nov 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13020 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	4th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2021
Country/Territory	China
City	Beijing
Period	29/10/21 → 1/11/21

Keywords

Face replacement
Style transfer
Traditional Chinese portrait

Access to Document

10.1007/978-3-030-88007-1_16

Cite this

Li, J., Wang, Q., Li, S., Zhong, Q., & Zhou, Q. (2021). Immersive Traditional Chinese Portrait Painting: Research on Style Transfer and Face Replacement. In H. Ma, L. Wang, C. Zhang, F. Wu, T. Tan, Y. Wang, J. Lai, & Y. Zhao (Eds.), Pattern Recognition and Computer Vision - 4th Chinese Conference, PRCV 2021, Proceedings (pp. 192-203). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13020 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-88007-1_16

Li, Jiayue ; Wang, Qing ; Li, Shiji et al. / Immersive Traditional Chinese Portrait Painting : Research on Style Transfer and Face Replacement. Pattern Recognition and Computer Vision - 4th Chinese Conference, PRCV 2021, Proceedings. editor / Huimin Ma ; Liang Wang ; Changshui Zhang ; Fei Wu ; Tieniu Tan ; Yaonan Wang ; Jianhuang Lai ; Yao Zhao. Springer Science and Business Media Deutschland GmbH, 2021. pp. 192-203 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{af6d3b230ee947bf9ff8ce59327a7d51,

title = "Immersive Traditional Chinese Portrait Painting: Research on Style Transfer and Face Replacement",

abstract = "Traditional Chinese portrait is popular all over the world because of its unique oriental charm. However, how to use neural network to express the aesthetic and feelings in instantiated Chinese portrait effectively is still a challenging problem. This paper proposes a Photo to Chinese Portrait method (P-CP) providing immersive traditional Chinese portrait painting experience. Our method can produce two groups intriguing pictures. One is Chinese Portrait Style Picture, the other is Immersive Chinese Portrait Picture. We pay attention to neural style transfer for traditional Chinese portrait for the first time, and have trained a fast feedforward generative network to extract the corresponding style. The generative network principle is explored to guide the style transfer adjustment in detail. Face replacement is added to form a more appealing stylized effect. We also solve the problems of color and light violation, and unnatural seam. We hope this work offers a deeper and immersive conversation between modern society and antiques, and provides a useful step towards related interdisciplinary areas.",

keywords = "Face replacement, Style transfer, Traditional Chinese portrait",

author = "Jiayue Li and Qing Wang and Shiji Li and Qiang Zhong and Qian Zhou",

note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 4th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2021 ; Conference date: 29-10-2021 Through 01-11-2021",

year = "2021",

doi = "10.1007/978-3-030-88007-1_16",

language = "英语",

isbn = "9783030880064",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "192--203",

editor = "Huimin Ma and Liang Wang and Changshui Zhang and Fei Wu and Tieniu Tan and Yaonan Wang and Jianhuang Lai and Yao Zhao",

booktitle = "Pattern Recognition and Computer Vision - 4th Chinese Conference, PRCV 2021, Proceedings",

}

Li, J, Wang, Q, Li, S, Zhong, Q & Zhou, Q 2021, Immersive Traditional Chinese Portrait Painting: Research on Style Transfer and Face Replacement. in H Ma, L Wang, C Zhang, F Wu, T Tan, Y Wang, J Lai & Y Zhao (eds), Pattern Recognition and Computer Vision - 4th Chinese Conference, PRCV 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13020 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 192-203, 4th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2021, Beijing, China, 29/10/21. https://doi.org/10.1007/978-3-030-88007-1_16

Immersive Traditional Chinese Portrait Painting: Research on Style Transfer and Face Replacement. / Li, Jiayue; Wang, Qing; Li, Shiji et al.
Pattern Recognition and Computer Vision - 4th Chinese Conference, PRCV 2021, Proceedings. ed. / Huimin Ma; Liang Wang; Changshui Zhang; Fei Wu; Tieniu Tan; Yaonan Wang; Jianhuang Lai; Yao Zhao. Springer Science and Business Media Deutschland GmbH, 2021. p. 192-203 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13020 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Immersive Traditional Chinese Portrait Painting

T2 - 4th Chinese Conference on Pattern Recognition and Computer Vision, PRCV 2021

AU - Li, Jiayue

AU - Wang, Qing

AU - Li, Shiji

AU - Zhong, Qiang

AU - Zhou, Qian

PY - 2021

Y1 - 2021

N2 - Traditional Chinese portrait is popular all over the world because of its unique oriental charm. However, how to use neural network to express the aesthetic and feelings in instantiated Chinese portrait effectively is still a challenging problem. This paper proposes a Photo to Chinese Portrait method (P-CP) providing immersive traditional Chinese portrait painting experience. Our method can produce two groups intriguing pictures. One is Chinese Portrait Style Picture, the other is Immersive Chinese Portrait Picture. We pay attention to neural style transfer for traditional Chinese portrait for the first time, and have trained a fast feedforward generative network to extract the corresponding style. The generative network principle is explored to guide the style transfer adjustment in detail. Face replacement is added to form a more appealing stylized effect. We also solve the problems of color and light violation, and unnatural seam. We hope this work offers a deeper and immersive conversation between modern society and antiques, and provides a useful step towards related interdisciplinary areas.

AB - Traditional Chinese portrait is popular all over the world because of its unique oriental charm. However, how to use neural network to express the aesthetic and feelings in instantiated Chinese portrait effectively is still a challenging problem. This paper proposes a Photo to Chinese Portrait method (P-CP) providing immersive traditional Chinese portrait painting experience. Our method can produce two groups intriguing pictures. One is Chinese Portrait Style Picture, the other is Immersive Chinese Portrait Picture. We pay attention to neural style transfer for traditional Chinese portrait for the first time, and have trained a fast feedforward generative network to extract the corresponding style. The generative network principle is explored to guide the style transfer adjustment in detail. Face replacement is added to form a more appealing stylized effect. We also solve the problems of color and light violation, and unnatural seam. We hope this work offers a deeper and immersive conversation between modern society and antiques, and provides a useful step towards related interdisciplinary areas.

KW - Face replacement

KW - Style transfer

KW - Traditional Chinese portrait

UR - http://www.scopus.com/inward/record.url?scp=85118224667&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-88007-1_16

DO - 10.1007/978-3-030-88007-1_16

M3 - 会议稿件

AN - SCOPUS:85118224667

SN - 9783030880064

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 192

EP - 203

BT - Pattern Recognition and Computer Vision - 4th Chinese Conference, PRCV 2021, Proceedings

A2 - Ma, Huimin

A2 - Wang, Liang

A2 - Zhang, Changshui

A2 - Wu, Fei

A2 - Tan, Tieniu

A2 - Wang, Yaonan

A2 - Lai, Jianhuang

A2 - Zhao, Yao

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 29 October 2021 through 1 November 2021

ER -

Li J, Wang Q, Li S, Zhong Q, Zhou Q. Immersive Traditional Chinese Portrait Painting: Research on Style Transfer and Face Replacement. In Ma H, Wang L, Zhang C, Wu F, Tan T, Wang Y, Lai J, Zhao Y, editors, Pattern Recognition and Computer Vision - 4th Chinese Conference, PRCV 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2021. p. 192-203. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-88007-1_16

Immersive Traditional Chinese Portrait Painting: Research on Style Transfer and Face Replacement

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this