Stylizing face images via multiple exemplars

Yibing Song, Linchao Bao, Shengfeng He, Qingxiong Yang, Ming Hsuan Yang

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

We address the problem of transferring the style of a headshot photo to face images. Existing methods using a single exemplar lead to inaccurate results when the exemplar does not contain sufficient stylized facial components for a given photo. In this work, we propose an algorithm to stylize face images using multiple exemplars containing different subjects in the same style. Patch correspondences between an input photo and multiple exemplars are established using a Markov Random Field (MRF), which enables accurate local energy transfer via Laplacian stacks. As image patches from multiple exemplars are used, the boundaries of facial components on the target image are inevitably inconsistent. The artifacts are removed by a post-processing step using an edge-preserving filter. Experimental results show that the proposed algorithm consistently produces visually pleasing results.

Original languageEnglish
Pages (from-to)135-145
Number of pages11
JournalComputer Vision and Image Understanding
Volume162
DOIs
Publication statusPublished - 2017 Sep

Fingerprint

Energy transfer
Processing

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition

Cite this

Song, Yibing ; Bao, Linchao ; He, Shengfeng ; Yang, Qingxiong ; Yang, Ming Hsuan. / Stylizing face images via multiple exemplars. In: Computer Vision and Image Understanding. 2017 ; Vol. 162. pp. 135-145.
@article{4e3c6268f6984392a941668745ecee71,
title = "Stylizing face images via multiple exemplars",
abstract = "We address the problem of transferring the style of a headshot photo to face images. Existing methods using a single exemplar lead to inaccurate results when the exemplar does not contain sufficient stylized facial components for a given photo. In this work, we propose an algorithm to stylize face images using multiple exemplars containing different subjects in the same style. Patch correspondences between an input photo and multiple exemplars are established using a Markov Random Field (MRF), which enables accurate local energy transfer via Laplacian stacks. As image patches from multiple exemplars are used, the boundaries of facial components on the target image are inevitably inconsistent. The artifacts are removed by a post-processing step using an edge-preserving filter. Experimental results show that the proposed algorithm consistently produces visually pleasing results.",
author = "Yibing Song and Linchao Bao and Shengfeng He and Qingxiong Yang and Yang, {Ming Hsuan}",
year = "2017",
month = "9",
doi = "10.1016/j.cviu.2017.08.009",
language = "English",
volume = "162",
pages = "135--145",
journal = "Computer Vision and Image Understanding",
issn = "1077-3142",
publisher = "Academic Press Inc.",

}

Stylizing face images via multiple exemplars. / Song, Yibing; Bao, Linchao; He, Shengfeng; Yang, Qingxiong; Yang, Ming Hsuan.

In: Computer Vision and Image Understanding, Vol. 162, 09.2017, p. 135-145.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Stylizing face images via multiple exemplars

AU - Song, Yibing

AU - Bao, Linchao

AU - He, Shengfeng

AU - Yang, Qingxiong

AU - Yang, Ming Hsuan

PY - 2017/9

Y1 - 2017/9

N2 - We address the problem of transferring the style of a headshot photo to face images. Existing methods using a single exemplar lead to inaccurate results when the exemplar does not contain sufficient stylized facial components for a given photo. In this work, we propose an algorithm to stylize face images using multiple exemplars containing different subjects in the same style. Patch correspondences between an input photo and multiple exemplars are established using a Markov Random Field (MRF), which enables accurate local energy transfer via Laplacian stacks. As image patches from multiple exemplars are used, the boundaries of facial components on the target image are inevitably inconsistent. The artifacts are removed by a post-processing step using an edge-preserving filter. Experimental results show that the proposed algorithm consistently produces visually pleasing results.

AB - We address the problem of transferring the style of a headshot photo to face images. Existing methods using a single exemplar lead to inaccurate results when the exemplar does not contain sufficient stylized facial components for a given photo. In this work, we propose an algorithm to stylize face images using multiple exemplars containing different subjects in the same style. Patch correspondences between an input photo and multiple exemplars are established using a Markov Random Field (MRF), which enables accurate local energy transfer via Laplacian stacks. As image patches from multiple exemplars are used, the boundaries of facial components on the target image are inevitably inconsistent. The artifacts are removed by a post-processing step using an edge-preserving filter. Experimental results show that the proposed algorithm consistently produces visually pleasing results.

UR - http://www.scopus.com/inward/record.url?scp=85028450105&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028450105&partnerID=8YFLogxK

U2 - 10.1016/j.cviu.2017.08.009

DO - 10.1016/j.cviu.2017.08.009

M3 - Article

AN - SCOPUS:85028450105

VL - 162

SP - 135

EP - 145

JO - Computer Vision and Image Understanding

JF - Computer Vision and Image Understanding

SN - 1077-3142

ER -