Extraction of non-binary blotch mattes

Mohamed Elgharib, François Pitié, Anil C. Kokaram

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

Automated blotch removal is important in film restoration and typically involves a detection/interpolation step. Current algorithms model the corruption as a binary mixture between the original, clean images and an opaque (dirt) field. This typically causes incomplete blotch removal that manifests as blotch haloes in reconstruction. This paper proposes a new approach by modeling the corruption as a continuous mixture between the two components and generating a solution using a Bayesian framework. We use novel priors, propose a computationally efficient scheme for implementation and our results show more complete blotch reconstruction.

Original languageEnglish
Title of host publicationProceedings - International Conference on Image Processing, ICIP
Pages2757-2760
Number of pages4
DOIs
Publication statusPublished - 1 Dec 2009
Externally publishedYes
Event2009 IEEE International Conference on Image Processing, ICIP 2009 - Cairo, Egypt
Duration: 7 Nov 200912 Nov 2009

Other

Other2009 IEEE International Conference on Image Processing, ICIP 2009
CountryEgypt
CityCairo
Period7/11/0912/11/09

Fingerprint

Binary mixtures
Restoration
Interpolation

Keywords

  • Bayesian matting
  • Blotch
  • Frames
  • Fusion
  • Matte
  • MRF
  • Non-binary
  • Opacity
  • Removal
  • Transparency

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Elgharib, M., Pitié, F., & Kokaram, A. C. (2009). Extraction of non-binary blotch mattes. In Proceedings - International Conference on Image Processing, ICIP (pp. 2757-2760). [5414134] https://doi.org/10.1109/ICIP.2009.5414134

Extraction of non-binary blotch mattes. / Elgharib, Mohamed; Pitié, François; Kokaram, Anil C.

Proceedings - International Conference on Image Processing, ICIP. 2009. p. 2757-2760 5414134.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Elgharib, M, Pitié, F & Kokaram, AC 2009, Extraction of non-binary blotch mattes. in Proceedings - International Conference on Image Processing, ICIP., 5414134, pp. 2757-2760, 2009 IEEE International Conference on Image Processing, ICIP 2009, Cairo, Egypt, 7/11/09. https://doi.org/10.1109/ICIP.2009.5414134
Elgharib M, Pitié F, Kokaram AC. Extraction of non-binary blotch mattes. In Proceedings - International Conference on Image Processing, ICIP. 2009. p. 2757-2760. 5414134 https://doi.org/10.1109/ICIP.2009.5414134
Elgharib, Mohamed ; Pitié, François ; Kokaram, Anil C. / Extraction of non-binary blotch mattes. Proceedings - International Conference on Image Processing, ICIP. 2009. pp. 2757-2760
@inproceedings{1da89b3be64944fcadecbde462544039,
title = "Extraction of non-binary blotch mattes",
abstract = "Automated blotch removal is important in film restoration and typically involves a detection/interpolation step. Current algorithms model the corruption as a binary mixture between the original, clean images and an opaque (dirt) field. This typically causes incomplete blotch removal that manifests as blotch haloes in reconstruction. This paper proposes a new approach by modeling the corruption as a continuous mixture between the two components and generating a solution using a Bayesian framework. We use novel priors, propose a computationally efficient scheme for implementation and our results show more complete blotch reconstruction.",
keywords = "Bayesian matting, Blotch, Frames, Fusion, Matte, MRF, Non-binary, Opacity, Removal, Transparency",
author = "Mohamed Elgharib and Fran{\cc}ois Piti{\'e} and Kokaram, {Anil C.}",
year = "2009",
month = "12",
day = "1",
doi = "10.1109/ICIP.2009.5414134",
language = "English",
isbn = "9781424456543",
pages = "2757--2760",
booktitle = "Proceedings - International Conference on Image Processing, ICIP",

}

TY - GEN

T1 - Extraction of non-binary blotch mattes

AU - Elgharib, Mohamed

AU - Pitié, François

AU - Kokaram, Anil C.

PY - 2009/12/1

Y1 - 2009/12/1

N2 - Automated blotch removal is important in film restoration and typically involves a detection/interpolation step. Current algorithms model the corruption as a binary mixture between the original, clean images and an opaque (dirt) field. This typically causes incomplete blotch removal that manifests as blotch haloes in reconstruction. This paper proposes a new approach by modeling the corruption as a continuous mixture between the two components and generating a solution using a Bayesian framework. We use novel priors, propose a computationally efficient scheme for implementation and our results show more complete blotch reconstruction.

AB - Automated blotch removal is important in film restoration and typically involves a detection/interpolation step. Current algorithms model the corruption as a binary mixture between the original, clean images and an opaque (dirt) field. This typically causes incomplete blotch removal that manifests as blotch haloes in reconstruction. This paper proposes a new approach by modeling the corruption as a continuous mixture between the two components and generating a solution using a Bayesian framework. We use novel priors, propose a computationally efficient scheme for implementation and our results show more complete blotch reconstruction.

KW - Bayesian matting

KW - Blotch

KW - Frames

KW - Fusion

KW - Matte

KW - MRF

KW - Non-binary

KW - Opacity

KW - Removal

KW - Transparency

UR - http://www.scopus.com/inward/record.url?scp=77951943717&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951943717&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2009.5414134

DO - 10.1109/ICIP.2009.5414134

M3 - Conference contribution

AN - SCOPUS:77951943717

SN - 9781424456543

SP - 2757

EP - 2760

BT - Proceedings - International Conference on Image Processing, ICIP

ER -