Background recovery from multiple images

Aditee Shrotre, Lina Karam

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

4 Citations (Scopus)

Abstract

In this paper we propose an algorithm to extract the background by removing unwanted objects from multiple images of a scene with varying illumination conditions captured by a stationary camera. The variations in illumination from scene to scene are due to the possible presence of different illumination sources and different foreground objects causing different shadows and reflections in each scene. While this causes the background to be non-stationary when considering pixel intensities, the algorithm exploits the fact that the background is static while the foreground is non-stationary in a given feature space. The extracted foreground regions are treated as holes and are filled from one of the available images where the background at the corresponding position is un-occluded. The background pixels for filling the holes are selected based on a cost function that attempts to maximize the naturalness and perceived quality of the reconstructed background.

Original languageEnglish (US)
Title of host publication2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013 - Proceedings
PublisherIEEE Computer Society
Pages135-140
Number of pages6
ISBN (Print)9781479916160
DOIs
StatePublished - 2013
Event2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013 - Napa, CA, United States
Duration: Aug 11 2013Aug 14 2013

Other

Other2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013
CountryUnited States
CityNapa, CA
Period8/11/138/14/13

Fingerprint

Lighting
Recovery
Pixels
Cost functions
Cameras

Keywords

  • Background recovery or background extraction
  • blending
  • feature descriptors
  • occlusion removal
  • patch matching

ASJC Scopus subject areas

  • Signal Processing

Cite this

Shrotre, A., & Karam, L. (2013). Background recovery from multiple images. In 2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013 - Proceedings (pp. 135-140). [6642579] IEEE Computer Society. https://doi.org/10.1109/DSP-SPE.2013.6642579

Background recovery from multiple images. / Shrotre, Aditee; Karam, Lina.

2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013 - Proceedings. IEEE Computer Society, 2013. p. 135-140 6642579.

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

Shrotre, A & Karam, L 2013, Background recovery from multiple images. in 2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013 - Proceedings., 6642579, IEEE Computer Society, pp. 135-140, 2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013, Napa, CA, United States, 8/11/13. https://doi.org/10.1109/DSP-SPE.2013.6642579
Shrotre A, Karam L. Background recovery from multiple images. In 2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013 - Proceedings. IEEE Computer Society. 2013. p. 135-140. 6642579 https://doi.org/10.1109/DSP-SPE.2013.6642579
Shrotre, Aditee ; Karam, Lina. / Background recovery from multiple images. 2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013 - Proceedings. IEEE Computer Society, 2013. pp. 135-140
@inproceedings{4e3003b8fbf6479794c112b24cfb07b3,
title = "Background recovery from multiple images",
abstract = "In this paper we propose an algorithm to extract the background by removing unwanted objects from multiple images of a scene with varying illumination conditions captured by a stationary camera. The variations in illumination from scene to scene are due to the possible presence of different illumination sources and different foreground objects causing different shadows and reflections in each scene. While this causes the background to be non-stationary when considering pixel intensities, the algorithm exploits the fact that the background is static while the foreground is non-stationary in a given feature space. The extracted foreground regions are treated as holes and are filled from one of the available images where the background at the corresponding position is un-occluded. The background pixels for filling the holes are selected based on a cost function that attempts to maximize the naturalness and perceived quality of the reconstructed background.",
keywords = "Background recovery or background extraction, blending, feature descriptors, occlusion removal, patch matching",
author = "Aditee Shrotre and Lina Karam",
year = "2013",
doi = "10.1109/DSP-SPE.2013.6642579",
language = "English (US)",
isbn = "9781479916160",
pages = "135--140",
booktitle = "2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013 - Proceedings",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - Background recovery from multiple images

AU - Shrotre, Aditee

AU - Karam, Lina

PY - 2013

Y1 - 2013

N2 - In this paper we propose an algorithm to extract the background by removing unwanted objects from multiple images of a scene with varying illumination conditions captured by a stationary camera. The variations in illumination from scene to scene are due to the possible presence of different illumination sources and different foreground objects causing different shadows and reflections in each scene. While this causes the background to be non-stationary when considering pixel intensities, the algorithm exploits the fact that the background is static while the foreground is non-stationary in a given feature space. The extracted foreground regions are treated as holes and are filled from one of the available images where the background at the corresponding position is un-occluded. The background pixels for filling the holes are selected based on a cost function that attempts to maximize the naturalness and perceived quality of the reconstructed background.

AB - In this paper we propose an algorithm to extract the background by removing unwanted objects from multiple images of a scene with varying illumination conditions captured by a stationary camera. The variations in illumination from scene to scene are due to the possible presence of different illumination sources and different foreground objects causing different shadows and reflections in each scene. While this causes the background to be non-stationary when considering pixel intensities, the algorithm exploits the fact that the background is static while the foreground is non-stationary in a given feature space. The extracted foreground regions are treated as holes and are filled from one of the available images where the background at the corresponding position is un-occluded. The background pixels for filling the holes are selected based on a cost function that attempts to maximize the naturalness and perceived quality of the reconstructed background.

KW - Background recovery or background extraction

KW - blending

KW - feature descriptors

KW - occlusion removal

KW - patch matching

UR - http://www.scopus.com/inward/record.url?scp=84890771631&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890771631&partnerID=8YFLogxK

U2 - 10.1109/DSP-SPE.2013.6642579

DO - 10.1109/DSP-SPE.2013.6642579

M3 - Conference contribution

SN - 9781479916160

SP - 135

EP - 140

BT - 2013 IEEE Digital Signal Processing and Signal Processing Education Meeting, DSP/SPE 2013 - Proceedings

PB - IEEE Computer Society

ER -