User-assisted image shadow removal

Han Gong, Darren Cosker

Research output: Contribution to journalArticle

  • 1 Citations

Abstract

This paper presents a novel user-aided method for texture-preserving shadow removal from single images requiring simple user input. Compared with the state-of-the-art, our algorithm offers the most flexible user interaction to date and produces more accurate and robust shadow removal under thorough quantitative evaluation. Shadow masks are first detected by analysing user specified shadow feature strokes. Sample intensity profiles with variable interval and length around the shadow boundary are detected next, which avoids artefacts raised from uneven boundaries. Texture noise in samples is then removed by applying local group bilateral filtering, and initial sparse shadow scales are estimated by fitting a piecewise curve to intensity samples. The remaining errors in estimated sparse scales are removed by local group smoothing. To relight the image, a dense scale field is produced by in-painting the sparse scales. Finally, a gradual colour correction is applied to remove artefacts due to image post-processing. Using state-of-the-art evaluation data, we quantitatively and qualitatively demonstrate our method to outperform current leading shadow removal methods.

LanguageEnglish
Pages19-27
Number of pages9
JournalImage and Vision Computing
Volume62
Early online date18 Apr 2017
DOIs
StatusPublished - 1 Jun 2017

Fingerprint

Textures
Painting
Masks
Color
Processing

Keywords

  • Colour correction
  • Curve fitting
  • Image shadow removal
  • Smoothing
  • User-assisted computer vision

ASJC Scopus subject areas

  • Electrical and Electronic Engineering

Cite this

User-assisted image shadow removal. / Gong, Han; Cosker, Darren.

In: Image and Vision Computing, Vol. 62, 01.06.2017, p. 19-27.

Research output: Contribution to journalArticle

@article{c1741322af094c349907d15560365ce2,
title = "User-assisted image shadow removal",
abstract = "This paper presents a novel user-aided method for texture-preserving shadow removal from single images requiring simple user input. Compared with the state-of-the-art, our algorithm offers the most flexible user interaction to date and produces more accurate and robust shadow removal under thorough quantitative evaluation. Shadow masks are first detected by analysing user specified shadow feature strokes. Sample intensity profiles with variable interval and length around the shadow boundary are detected next, which avoids artefacts raised from uneven boundaries. Texture noise in samples is then removed by applying local group bilateral filtering, and initial sparse shadow scales are estimated by fitting a piecewise curve to intensity samples. The remaining errors in estimated sparse scales are removed by local group smoothing. To relight the image, a dense scale field is produced by in-painting the sparse scales. Finally, a gradual colour correction is applied to remove artefacts due to image post-processing. Using state-of-the-art evaluation data, we quantitatively and qualitatively demonstrate our method to outperform current leading shadow removal methods.",
keywords = "Colour correction, Curve fitting, Image shadow removal, Smoothing, User-assisted computer vision",
author = "Han Gong and Darren Cosker",
year = "2017",
month = "6",
day = "1",
doi = "10.1016/j.imavis.2017.04.001",
language = "English",
volume = "62",
pages = "19--27",
journal = "Image and Vision Computing",
issn = "0262-8856",
publisher = "Elsevier",

}

TY - JOUR

T1 - User-assisted image shadow removal

AU - Gong,Han

AU - Cosker,Darren

PY - 2017/6/1

Y1 - 2017/6/1

N2 - This paper presents a novel user-aided method for texture-preserving shadow removal from single images requiring simple user input. Compared with the state-of-the-art, our algorithm offers the most flexible user interaction to date and produces more accurate and robust shadow removal under thorough quantitative evaluation. Shadow masks are first detected by analysing user specified shadow feature strokes. Sample intensity profiles with variable interval and length around the shadow boundary are detected next, which avoids artefacts raised from uneven boundaries. Texture noise in samples is then removed by applying local group bilateral filtering, and initial sparse shadow scales are estimated by fitting a piecewise curve to intensity samples. The remaining errors in estimated sparse scales are removed by local group smoothing. To relight the image, a dense scale field is produced by in-painting the sparse scales. Finally, a gradual colour correction is applied to remove artefacts due to image post-processing. Using state-of-the-art evaluation data, we quantitatively and qualitatively demonstrate our method to outperform current leading shadow removal methods.

AB - This paper presents a novel user-aided method for texture-preserving shadow removal from single images requiring simple user input. Compared with the state-of-the-art, our algorithm offers the most flexible user interaction to date and produces more accurate and robust shadow removal under thorough quantitative evaluation. Shadow masks are first detected by analysing user specified shadow feature strokes. Sample intensity profiles with variable interval and length around the shadow boundary are detected next, which avoids artefacts raised from uneven boundaries. Texture noise in samples is then removed by applying local group bilateral filtering, and initial sparse shadow scales are estimated by fitting a piecewise curve to intensity samples. The remaining errors in estimated sparse scales are removed by local group smoothing. To relight the image, a dense scale field is produced by in-painting the sparse scales. Finally, a gradual colour correction is applied to remove artefacts due to image post-processing. Using state-of-the-art evaluation data, we quantitatively and qualitatively demonstrate our method to outperform current leading shadow removal methods.

KW - Colour correction

KW - Curve fitting

KW - Image shadow removal

KW - Smoothing

KW - User-assisted computer vision

UR - http://www.scopus.com/inward/record.url?scp=85018877931&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1016/j.imavis.2017.04.001

U2 - 10.1016/j.imavis.2017.04.001

DO - 10.1016/j.imavis.2017.04.001

M3 - Article

VL - 62

SP - 19

EP - 27

JO - Image and Vision Computing

T2 - Image and Vision Computing

JF - Image and Vision Computing

SN - 0262-8856

ER -