Upsampling and denoising of depth maps via joint-segmentation

Miguel Tallón, S. Derin Babacan, Javier Mateos, Minh N Do, Rafael Molina, Aggelos K. Katsaggelos

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

Abstract

The recent development of low-cost and fast time-of-flight cameras enabled measuring depth information at video frame rates. Although these cameras provide invaluable information for many 3D applications, their imaging capabilities are very limited both in terms of resolution and noise level. In this paper, we present a novel method for obtaining a high resolution depth map from a pair of a low resolution depth map and a corresponding high resolution color image. The proposed method exploits the correlation between the objects present in the color and depth map images via joint segmentation, which is then used to increase the resolution and remove noise via estimating conditional modes. Regions with inconsistent color and depth information are detected and corrected with our algorithm for increased robustness. Experimental results in terms of image quality and running times demonstrate the high performance of the method.

Original languageEnglish (US)
Title of host publicationProceedings of the 20th European Signal Processing Conference, EUSIPCO 2012
Pages245-249
Number of pages5
StatePublished - 2012
Event20th European Signal Processing Conference, EUSIPCO 2012 - Bucharest
Duration: Aug 27 2012Aug 31 2012

Other

Other20th European Signal Processing Conference, EUSIPCO 2012
CityBucharest
Period8/27/128/31/12

Fingerprint

Color
Cameras
Image quality
Imaging techniques
Costs

Keywords

  • color segmentation
  • depth enhancement
  • multisensor image fusion
  • Time-of-flight cameras

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Tallón, M., Derin Babacan, S., Mateos, J., Do, M. N., Molina, R., & Katsaggelos, A. K. (2012). Upsampling and denoising of depth maps via joint-segmentation. In Proceedings of the 20th European Signal Processing Conference, EUSIPCO 2012 (pp. 245-249). [6334071]

Upsampling and denoising of depth maps via joint-segmentation. / Tallón, Miguel; Derin Babacan, S.; Mateos, Javier; Do, Minh N; Molina, Rafael; Katsaggelos, Aggelos K.

Proceedings of the 20th European Signal Processing Conference, EUSIPCO 2012. 2012. p. 245-249 6334071.

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

Tallón, M, Derin Babacan, S, Mateos, J, Do, MN, Molina, R & Katsaggelos, AK 2012, Upsampling and denoising of depth maps via joint-segmentation. in Proceedings of the 20th European Signal Processing Conference, EUSIPCO 2012., 6334071, pp. 245-249, 20th European Signal Processing Conference, EUSIPCO 2012, Bucharest, 8/27/12.
Tallón M, Derin Babacan S, Mateos J, Do MN, Molina R, Katsaggelos AK. Upsampling and denoising of depth maps via joint-segmentation. In Proceedings of the 20th European Signal Processing Conference, EUSIPCO 2012. 2012. p. 245-249. 6334071
Tallón, Miguel ; Derin Babacan, S. ; Mateos, Javier ; Do, Minh N ; Molina, Rafael ; Katsaggelos, Aggelos K. / Upsampling and denoising of depth maps via joint-segmentation. Proceedings of the 20th European Signal Processing Conference, EUSIPCO 2012. 2012. pp. 245-249
@inproceedings{6dcf146c6b50431db904fbc92d15c22f,
title = "Upsampling and denoising of depth maps via joint-segmentation",
abstract = "The recent development of low-cost and fast time-of-flight cameras enabled measuring depth information at video frame rates. Although these cameras provide invaluable information for many 3D applications, their imaging capabilities are very limited both in terms of resolution and noise level. In this paper, we present a novel method for obtaining a high resolution depth map from a pair of a low resolution depth map and a corresponding high resolution color image. The proposed method exploits the correlation between the objects present in the color and depth map images via joint segmentation, which is then used to increase the resolution and remove noise via estimating conditional modes. Regions with inconsistent color and depth information are detected and corrected with our algorithm for increased robustness. Experimental results in terms of image quality and running times demonstrate the high performance of the method.",
keywords = "color segmentation, depth enhancement, multisensor image fusion, Time-of-flight cameras",
author = "Miguel Tall{\'o}n and {Derin Babacan}, S. and Javier Mateos and Do, {Minh N} and Rafael Molina and Katsaggelos, {Aggelos K.}",
year = "2012",
language = "English (US)",
isbn = "9781467310680",
pages = "245--249",
booktitle = "Proceedings of the 20th European Signal Processing Conference, EUSIPCO 2012",

}

TY - GEN

T1 - Upsampling and denoising of depth maps via joint-segmentation

AU - Tallón, Miguel

AU - Derin Babacan, S.

AU - Mateos, Javier

AU - Do, Minh N

AU - Molina, Rafael

AU - Katsaggelos, Aggelos K.

PY - 2012

Y1 - 2012

N2 - The recent development of low-cost and fast time-of-flight cameras enabled measuring depth information at video frame rates. Although these cameras provide invaluable information for many 3D applications, their imaging capabilities are very limited both in terms of resolution and noise level. In this paper, we present a novel method for obtaining a high resolution depth map from a pair of a low resolution depth map and a corresponding high resolution color image. The proposed method exploits the correlation between the objects present in the color and depth map images via joint segmentation, which is then used to increase the resolution and remove noise via estimating conditional modes. Regions with inconsistent color and depth information are detected and corrected with our algorithm for increased robustness. Experimental results in terms of image quality and running times demonstrate the high performance of the method.

AB - The recent development of low-cost and fast time-of-flight cameras enabled measuring depth information at video frame rates. Although these cameras provide invaluable information for many 3D applications, their imaging capabilities are very limited both in terms of resolution and noise level. In this paper, we present a novel method for obtaining a high resolution depth map from a pair of a low resolution depth map and a corresponding high resolution color image. The proposed method exploits the correlation between the objects present in the color and depth map images via joint segmentation, which is then used to increase the resolution and remove noise via estimating conditional modes. Regions with inconsistent color and depth information are detected and corrected with our algorithm for increased robustness. Experimental results in terms of image quality and running times demonstrate the high performance of the method.

KW - color segmentation

KW - depth enhancement

KW - multisensor image fusion

KW - Time-of-flight cameras

UR - http://www.scopus.com/inward/record.url?scp=84869833634&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84869833634&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9781467310680

SP - 245

EP - 249

BT - Proceedings of the 20th European Signal Processing Conference, EUSIPCO 2012

ER -