Vectorization based color transfer for portrait images

This paper introduces a method for transferring colors between portrait images. Using a trained neural network to extract facial mask, we vectorize each image with a set of sparse diffusion curves to encode the low-frequency colors, and use the Laplacian of residual colors to represent the high-freq...

Full description

Saved in:
Bibliographic Details
Main Authors: Fu, Qian, He, Ying, Hou, Fei, Zhang, Juyong, Zeng, Anxiang, Liu, Yong-Jin
Other Authors: School of Computer Science and Engineering
Format: Article
Language:English
Published: 2021
Subjects:
Online Access:https://hdl.handle.net/10356/151525
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-151525
record_format dspace
spelling sg-ntu-dr.10356-1515252021-06-18T03:46:37Z Vectorization based color transfer for portrait images Fu, Qian He, Ying Hou, Fei Zhang, Juyong Zeng, Anxiang Liu, Yong-Jin School of Computer Science and Engineering NTU-Alibaba Joint Research Institute Engineering::Computer science and engineering Color Transfer Portrait Images This paper introduces a method for transferring colors between portrait images. Using a trained neural network to extract facial mask, we vectorize each image with a set of sparse diffusion curves to encode the low-frequency colors, and use the Laplacian of residual colors to represent the high-frequency details. Then we apply optimal mass transport to transfer the boundary colors between the diffusion curves of the source and reference images. Finally, the original or modified Laplacians of colors are added to the transferred diffusion curve image. Unlike the existing methods that either require 3D information or assume the source and reference images have similar poses and dense correspondence, our method is computationally efficient and flexible, which can work for portrait images with large pose and color differences. Ministry of Education (MOE) This project was partially supported by Singapore Ministry of Education Grant RG26/17, NTU-Alibaba Joint Research Institute, Singapore, National Natural Science Foundation of China Grants (61872347, 61672481, 61725204, and U1736220), Special Plan for the Development of Distinguished Young Scientists of ISCAS, China (Y8RC535018), Youth Innovation Promotion Association CAS, China (No. 2018495) and the Royal Society-Newton Advanced Fellowship, China (NA150431). 2021-06-18T03:46:37Z 2021-06-18T03:46:37Z 2019 Journal Article Fu, Q., He, Y., Hou, F., Zhang, J., Zeng, A. & Liu, Y. (2019). Vectorization based color transfer for portrait images. Computer Aided Design, 115, 111-121. https://dx.doi.org/10.1016/j.cad.2019.05.005 0010-4485 https://hdl.handle.net/10356/151525 10.1016/j.cad.2019.05.005 2-s2.0-85066285094 115 111 121 en RG26/17 Computer Aided Design © 2019 Elsevier Ltd. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Computer science and engineering
Color Transfer
Portrait Images
spellingShingle Engineering::Computer science and engineering
Color Transfer
Portrait Images
Fu, Qian
He, Ying
Hou, Fei
Zhang, Juyong
Zeng, Anxiang
Liu, Yong-Jin
Vectorization based color transfer for portrait images
description This paper introduces a method for transferring colors between portrait images. Using a trained neural network to extract facial mask, we vectorize each image with a set of sparse diffusion curves to encode the low-frequency colors, and use the Laplacian of residual colors to represent the high-frequency details. Then we apply optimal mass transport to transfer the boundary colors between the diffusion curves of the source and reference images. Finally, the original or modified Laplacians of colors are added to the transferred diffusion curve image. Unlike the existing methods that either require 3D information or assume the source and reference images have similar poses and dense correspondence, our method is computationally efficient and flexible, which can work for portrait images with large pose and color differences.
author2 School of Computer Science and Engineering
author_facet School of Computer Science and Engineering
Fu, Qian
He, Ying
Hou, Fei
Zhang, Juyong
Zeng, Anxiang
Liu, Yong-Jin
format Article
author Fu, Qian
He, Ying
Hou, Fei
Zhang, Juyong
Zeng, Anxiang
Liu, Yong-Jin
author_sort Fu, Qian
title Vectorization based color transfer for portrait images
title_short Vectorization based color transfer for portrait images
title_full Vectorization based color transfer for portrait images
title_fullStr Vectorization based color transfer for portrait images
title_full_unstemmed Vectorization based color transfer for portrait images
title_sort vectorization based color transfer for portrait images
publishDate 2021
url https://hdl.handle.net/10356/151525
_version_ 1703971231770869760