Restoration of Depth and Intensity Images using a Graph Laplacian Regularization

Abderrahim Halimi; Peter Connolly; Ximing Ren; Yoann Altmann; Istvan Gyongy; Robert K. Henderson; Stephen McLaughlin; Gerald Stuart Buller

doi:10.1109/CAMSAP.2017.8313136

Restoration of Depth and Intensity Images using a Graph Laplacian Regularization

Abderrahim Halimi, Peter Connolly, Ximing Ren, Yoann Altmann, Istvan Gyongy, Robert K. Henderson, Stephen McLaughlin, Gerald Stuart Buller

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

108 Downloads (Pure)

Abstract

This paper presents a new algorithm for the joint restoration of depth and intensity (DI) images constructed using a gated SPAD-array imaging system. The three dimensional (3D) data consists of two spatial dimensions and one temporal dimension, and contains photon counts (i.e., histograms). The algorithm is based on two steps: (i) construction of a graph connecting patches of pixels with similar temporal responses, and (ii) estimation of the DI values for pixels belonging to homogeneous spatial classes. The first step is achieved by building a graph representation of the 3D data, while giving a special attention to the computational complexity of the algorithm.
The second step is achieved using a Fisher scoring gradient descent algorithm while accounting for the data statistics and the Laplacian regularization term. Results on laboratory data show the benefit of the proposed strategy that improves the quality of the estimated DI images.

Original language	English
Title of host publication	2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)
Publisher	IEEE
ISBN (Electronic)	9781538612514
DOIs	https://doi.org/10.1109/CAMSAP.2017.8313136
Publication status	Published - 12 Mar 2018
Event	7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing 2017 - Curacao, Curaçao Duration: 10 Dec 2017 → 13 Dec 2017 http://www.cs.huji.ac.il/conferences/CAMSAP17/

Conference

Conference	7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing 2017
Abbreviated title	CAMSAP 2017
Country/Territory	Curaçao
City	Curacao
Period	10/12/17 → 13/12/17
Internet address	http://www.cs.huji.ac.il/conferences/CAMSAP17/

Access to Document

10.1109/CAMSAP.2017.8313136

Download pdf
© 2018 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript, 3.22 MB

Cite this

Halimi, A., Connolly, P., Ren, X., Altmann, Y., Gyongy, I., Henderson, R. K., McLaughlin, S., & Buller, G. S. (2018). Restoration of Depth and Intensity Images using a Graph Laplacian Regularization. In 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP) Article 8313136 IEEE. https://doi.org/10.1109/CAMSAP.2017.8313136

@inproceedings{9cf62d4a72254a9cbfe65e2d46a50ab4,

title = "Restoration of Depth and Intensity Images using a Graph Laplacian Regularization",

abstract = "This paper presents a new algorithm for the joint restoration of depth and intensity (DI) images constructed using a gated SPAD-array imaging system. The three dimensional (3D) data consists of two spatial dimensions and one temporal dimension, and contains photon counts (i.e., histograms). The algorithm is based on two steps: (i) construction of a graph connecting patches of pixels with similar temporal responses, and (ii) estimation of the DI values for pixels belonging to homogeneous spatial classes. The first step is achieved by building a graph representation of the 3D data, while giving a special attention to the computational complexity of the algorithm. The second step is achieved using a Fisher scoring gradient descent algorithm while accounting for the data statistics and the Laplacian regularization term. Results on laboratory data show the benefit of the proposed strategy that improves the quality of the estimated DI images. ",

author = "Abderrahim Halimi and Peter Connolly and Ximing Ren and Yoann Altmann and Istvan Gyongy and Henderson, {Robert K.} and Stephen McLaughlin and Buller, {Gerald Stuart}",

year = "2018",

month = mar,

day = "12",

doi = "10.1109/CAMSAP.2017.8313136",

language = "English",

booktitle = "2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)",

publisher = "IEEE",

address = "United States",

note = "7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing 2017, CAMSAP 2017 ; Conference date: 10-12-2017 Through 13-12-2017",

url = "http://www.cs.huji.ac.il/conferences/CAMSAP17/",

}

Halimi, A , Connolly, P, Ren, X, Altmann, Y, Gyongy, I, Henderson, RK, McLaughlin, S & Buller, GS 2018, Restoration of Depth and Intensity Images using a Graph Laplacian Regularization. in 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)., 8313136, IEEE, 7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing 2017, Curacao, Curaçao, 10/12/17. https://doi.org/10.1109/CAMSAP.2017.8313136

TY - GEN

T1 - Restoration of Depth and Intensity Images using a Graph Laplacian Regularization

AU - Halimi, Abderrahim

AU - Connolly, Peter

AU - Ren, Ximing

AU - Altmann, Yoann

AU - Gyongy, Istvan

AU - Henderson, Robert K.

AU - McLaughlin, Stephen

AU - Buller, Gerald Stuart

PY - 2018/3/12

Y1 - 2018/3/12

N2 - This paper presents a new algorithm for the joint restoration of depth and intensity (DI) images constructed using a gated SPAD-array imaging system. The three dimensional (3D) data consists of two spatial dimensions and one temporal dimension, and contains photon counts (i.e., histograms). The algorithm is based on two steps: (i) construction of a graph connecting patches of pixels with similar temporal responses, and (ii) estimation of the DI values for pixels belonging to homogeneous spatial classes. The first step is achieved by building a graph representation of the 3D data, while giving a special attention to the computational complexity of the algorithm. The second step is achieved using a Fisher scoring gradient descent algorithm while accounting for the data statistics and the Laplacian regularization term. Results on laboratory data show the benefit of the proposed strategy that improves the quality of the estimated DI images.

AB - This paper presents a new algorithm for the joint restoration of depth and intensity (DI) images constructed using a gated SPAD-array imaging system. The three dimensional (3D) data consists of two spatial dimensions and one temporal dimension, and contains photon counts (i.e., histograms). The algorithm is based on two steps: (i) construction of a graph connecting patches of pixels with similar temporal responses, and (ii) estimation of the DI values for pixels belonging to homogeneous spatial classes. The first step is achieved by building a graph representation of the 3D data, while giving a special attention to the computational complexity of the algorithm. The second step is achieved using a Fisher scoring gradient descent algorithm while accounting for the data statistics and the Laplacian regularization term. Results on laboratory data show the benefit of the proposed strategy that improves the quality of the estimated DI images.

U2 - 10.1109/CAMSAP.2017.8313136

DO - 10.1109/CAMSAP.2017.8313136

M3 - Conference contribution

BT - 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP)

PB - IEEE

T2 - 7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing 2017

Y2 - 10 December 2017 through 13 December 2017

ER -

Restoration of Depth and Intensity Images using a Graph Laplacian Regularization

Abstract

Conference

Access to Document

Fingerprint

Cite this