Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Robust image-to-image color transfer using optimal inlier maximization

Oskarsson, Magnus LU orcid (2021) 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021 In IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops p.786-795
Abstract

In this paper we target the color transfer estimation problem, when we have pixel-to-pixel correspondences. We present a feature-based method, that robustly fits color transforms to data containing gross outliers. Our solution is based on an optimal inlier maximization algorithm that maximizes the number of inliers in polynomial time. We introduce a simple feature detector and descriptor based on the structure tensor that gives the means for reliable matching of the color distributions in two images. Using combinatorial methods from optimization theory and a number of new minimal solvers, we can enumerate all possible stationary points to the inlier maximization problem. In order for our method to be tractable we use a decoupling of the... (More)

In this paper we target the color transfer estimation problem, when we have pixel-to-pixel correspondences. We present a feature-based method, that robustly fits color transforms to data containing gross outliers. Our solution is based on an optimal inlier maximization algorithm that maximizes the number of inliers in polynomial time. We introduce a simple feature detector and descriptor based on the structure tensor that gives the means for reliable matching of the color distributions in two images. Using combinatorial methods from optimization theory and a number of new minimal solvers, we can enumerate all possible stationary points to the inlier maximization problem. In order for our method to be tractable we use a decoupling of the intensity and color direction for a given RGB-vector. This enables the intensity transformation and the color direction transformation to be handled separately. Our method gives results comparable to state-of-the-art methods in the presence of little outliers, and large improvement for moderate or large amounts of outliers in the data. The proposed method has been tested in a number of imaging applications.

(Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
series title
IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
pages
10 pages
publisher
IEEE Computer Society
conference name
2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
conference location
Virtual, Online, United States
conference dates
2021-06-19 - 2021-06-25
external identifiers
  • scopus:85116054421
ISSN
2160-7508
2160-7516
ISBN
9781665448994
DOI
10.1109/CVPRW53098.2021.00088
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2021 IEEE.
id
ea835a45-3d59-4561-967b-37036ad265e0
date added to LUP
2021-10-19 15:24:14
date last changed
2024-09-08 02:26:50
@inproceedings{ea835a45-3d59-4561-967b-37036ad265e0,
  abstract     = {{<p>In this paper we target the color transfer estimation problem, when we have pixel-to-pixel correspondences. We present a feature-based method, that robustly fits color transforms to data containing gross outliers. Our solution is based on an optimal inlier maximization algorithm that maximizes the number of inliers in polynomial time. We introduce a simple feature detector and descriptor based on the structure tensor that gives the means for reliable matching of the color distributions in two images. Using combinatorial methods from optimization theory and a number of new minimal solvers, we can enumerate all possible stationary points to the inlier maximization problem. In order for our method to be tractable we use a decoupling of the intensity and color direction for a given RGB-vector. This enables the intensity transformation and the color direction transformation to be handled separately. Our method gives results comparable to state-of-the-art methods in the presence of little outliers, and large improvement for moderate or large amounts of outliers in the data. The proposed method has been tested in a number of imaging applications.</p>}},
  author       = {{Oskarsson, Magnus}},
  booktitle    = {{Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021}},
  isbn         = {{9781665448994}},
  issn         = {{2160-7508}},
  language     = {{eng}},
  pages        = {{786--795}},
  publisher    = {{IEEE Computer Society}},
  series       = {{IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops}},
  title        = {{Robust image-to-image color transfer using optimal inlier maximization}},
  url          = {{http://dx.doi.org/10.1109/CVPRW53098.2021.00088}},
  doi          = {{10.1109/CVPRW53098.2021.00088}},
  year         = {{2021}},
}