User-aided single image shadow removal

Han Gong, Darren Cosker, Chuan Li, Matthew Brown

Research output: Contribution to conferencePaper

  • 7 Citations

Abstract

This paper presents a novel user-aided method for texture preserving shadow removal from single images which only requires simple user input. Compared with the state-of-the-art, our algorithm addresses limitations in uneven shadow boundary processing and umbra recovery. We first detect an initial shadow boundary by growing a user specified shadow outline on an illumination-sensitive image. Interval-variable intensity sampling is introduced to avoid artefacts raised from uneven boundaries. We extract the initial scale field by applying local group intensity spline fittings around the shadow boundary. Bad intensity samples are replaced by their nearest alternatives based on a log-normal probability distribution of fitting errors. Finally, we use a gradual colour transfer to correct post-processing artefacts such as gamma correction and lossy compression. Compared with state-of-the-art methods, we offer highly user-friendly interaction, produce improved umbra recovery and improved processing given uneven shadow boundaries.

Conference

Conference2013 IEEE International Conference on Multimedia and Expo (ICME)
CountryUSA United States
CitySan Jose, California
Period14/07/1318/07/13

Fingerprint

Processing
Recovery
Splines
Probability distributions
Textures
Lighting
Sampling
Color

Keywords

  • arrays
  • estimation
  • image color analysis
  • image segmentation
  • lighting
  • optimization
  • splines (mathematics)
  • shadow removal
  • single image
  • user-aided

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Gong, H., Cosker, D., Li, C., & Brown, M. (2013). User-aided single image shadow removal. 1-6. Paper presented at 2013 IEEE International Conference on Multimedia and Expo (ICME), San Jose, California, USA United States.DOI: 10.1109/ICME.2013.6607463

User-aided single image shadow removal. / Gong, Han; Cosker, Darren; Li, Chuan; Brown, Matthew.

2013. 1-6 Paper presented at 2013 IEEE International Conference on Multimedia and Expo (ICME), San Jose, California, USA United States.

Research output: Contribution to conferencePaper

Gong, H, Cosker, D, Li, C & Brown, M 2013, 'User-aided single image shadow removal' Paper presented at 2013 IEEE International Conference on Multimedia and Expo (ICME), San Jose, California, USA United States, 14/07/13 - 18/07/13, pp. 1-6. DOI: 10.1109/ICME.2013.6607463
Gong H, Cosker D, Li C, Brown M. User-aided single image shadow removal. 2013. Paper presented at 2013 IEEE International Conference on Multimedia and Expo (ICME), San Jose, California, USA United States. Available from, DOI: 10.1109/ICME.2013.6607463
Gong, Han ; Cosker, Darren ; Li, Chuan ; Brown, Matthew. / User-aided single image shadow removal. Paper presented at 2013 IEEE International Conference on Multimedia and Expo (ICME), San Jose, California, USA United States.6 p.
@conference{1795e8c47edb4e87bc009fceebec0916,
title = "User-aided single image shadow removal",
abstract = "This paper presents a novel user-aided method for texture preserving shadow removal from single images which only requires simple user input. Compared with the state-of-the-art, our algorithm addresses limitations in uneven shadow boundary processing and umbra recovery. We first detect an initial shadow boundary by growing a user specified shadow outline on an illumination-sensitive image. Interval-variable intensity sampling is introduced to avoid artefacts raised from uneven boundaries. We extract the initial scale field by applying local group intensity spline fittings around the shadow boundary. Bad intensity samples are replaced by their nearest alternatives based on a log-normal probability distribution of fitting errors. Finally, we use a gradual colour transfer to correct post-processing artefacts such as gamma correction and lossy compression. Compared with state-of-the-art methods, we offer highly user-friendly interaction, produce improved umbra recovery and improved processing given uneven shadow boundaries.",
keywords = "arrays, estimation, image color analysis, image segmentation, lighting, optimization, splines (mathematics), shadow removal, single image, user-aided",
author = "Han Gong and Darren Cosker and Chuan Li and Matthew Brown",
year = "2013",
doi = "10.1109/ICME.2013.6607463",
language = "English",
pages = "1--6",
note = "2013 IEEE International Conference on Multimedia and Expo (ICME) ; Conference date: 14-07-2013 Through 18-07-2013",

}

TY - CONF

T1 - User-aided single image shadow removal

AU - Gong,Han

AU - Cosker,Darren

AU - Li,Chuan

AU - Brown,Matthew

PY - 2013

Y1 - 2013

N2 - This paper presents a novel user-aided method for texture preserving shadow removal from single images which only requires simple user input. Compared with the state-of-the-art, our algorithm addresses limitations in uneven shadow boundary processing and umbra recovery. We first detect an initial shadow boundary by growing a user specified shadow outline on an illumination-sensitive image. Interval-variable intensity sampling is introduced to avoid artefacts raised from uneven boundaries. We extract the initial scale field by applying local group intensity spline fittings around the shadow boundary. Bad intensity samples are replaced by their nearest alternatives based on a log-normal probability distribution of fitting errors. Finally, we use a gradual colour transfer to correct post-processing artefacts such as gamma correction and lossy compression. Compared with state-of-the-art methods, we offer highly user-friendly interaction, produce improved umbra recovery and improved processing given uneven shadow boundaries.

AB - This paper presents a novel user-aided method for texture preserving shadow removal from single images which only requires simple user input. Compared with the state-of-the-art, our algorithm addresses limitations in uneven shadow boundary processing and umbra recovery. We first detect an initial shadow boundary by growing a user specified shadow outline on an illumination-sensitive image. Interval-variable intensity sampling is introduced to avoid artefacts raised from uneven boundaries. We extract the initial scale field by applying local group intensity spline fittings around the shadow boundary. Bad intensity samples are replaced by their nearest alternatives based on a log-normal probability distribution of fitting errors. Finally, we use a gradual colour transfer to correct post-processing artefacts such as gamma correction and lossy compression. Compared with state-of-the-art methods, we offer highly user-friendly interaction, produce improved umbra recovery and improved processing given uneven shadow boundaries.

KW - arrays

KW - estimation

KW - image color analysis

KW - image segmentation

KW - lighting

KW - optimization

KW - splines (mathematics)

KW - shadow removal

KW - single image

KW - user-aided

UR - http://dx.doi.org/10.1109/ICME.2013.6607463

UR - http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6607463

U2 - 10.1109/ICME.2013.6607463

DO - 10.1109/ICME.2013.6607463

M3 - Paper

SP - 1

EP - 6

ER -