Sparse Signal Reconstruction with Multiple Side Information using Adaptive Weights for Multiview Sources

Huynh Van Luong, Jürgen Seiler, André Kaup, Søren Forchhammer

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

155 Downloads (Pure)

Abstract

This work considers reconstructing a target signal in a context ofdistributed sparse sources. We propose an efficient reconstruction algorithmwith the aid of other given sources as multiple side information (SI). Theproposed algorithm takes advantage of compressive sensing (CS) with SI andadaptive weights by solving a proposed weighted $n$-$\ell_{1}$ minimization.The proposed algorithm computes the adaptive weights in two levels, first eachindividual intra-SI and then inter-SI weights are iteratively updated at everyreconstructed iteration. This two-level optimization leads the proposedreconstruction algorithm with multiple SI using adaptive weights (RAMSIA) torobustly exploit the multiple SIs with different qualities. We experimentallyperform our algorithm on generated sparse signals and also correlated featurehistograms as multiview sparse sources from a multiview image database. Theresults show that RAMSIA significantly outperforms both classical CS and CSwith single SI, and RAMSIA with higher number of SIs gained more than the onewith smaller number of SIs.
Original languageEnglish
Title of host publicationProceedings of IEEE International Conference on Image Processing 2016
PublisherIEEE
Publication date2016
Pages2534-2538
ISBN (Print)9781467399616
DOIs
Publication statusPublished - 2016
Event23rd IEEE International Conference on Image Processing - Phoenix Convention Centre, Phoenix, AZ, United States
Duration: 25 Sep 201628 Sep 2016

Conference

Conference23rd IEEE International Conference on Image Processing
LocationPhoenix Convention Centre
CountryUnited States
CityPhoenix, AZ
Period25/09/201628/09/2016

Keywords

  • Side information
  • math.OC
  • Compressive sensing
  • Sparse signal
  • n-`1 minimization
  • Adaptive weights

Cite this

Luong, H. V., Seiler, J., Kaup, A., & Forchhammer, S. (2016). Sparse Signal Reconstruction with Multiple Side Information using Adaptive Weights for Multiview Sources. In Proceedings of IEEE International Conference on Image Processing 2016 (pp. 2534-2538). IEEE. https://doi.org/10.1109/ICIP.2016.7532816
Luong, Huynh Van ; Seiler, Jürgen ; Kaup, André ; Forchhammer, Søren. / Sparse Signal Reconstruction with Multiple Side Information using Adaptive Weights for Multiview Sources. Proceedings of IEEE International Conference on Image Processing 2016. IEEE, 2016. pp. 2534-2538
@inproceedings{d1463766673c42d0891478ed4494343e,
title = "Sparse Signal Reconstruction with Multiple Side Information using Adaptive Weights for Multiview Sources",
abstract = "This work considers reconstructing a target signal in a context ofdistributed sparse sources. We propose an efficient reconstruction algorithmwith the aid of other given sources as multiple side information (SI). Theproposed algorithm takes advantage of compressive sensing (CS) with SI andadaptive weights by solving a proposed weighted $n$-$\ell_{1}$ minimization.The proposed algorithm computes the adaptive weights in two levels, first eachindividual intra-SI and then inter-SI weights are iteratively updated at everyreconstructed iteration. This two-level optimization leads the proposedreconstruction algorithm with multiple SI using adaptive weights (RAMSIA) torobustly exploit the multiple SIs with different qualities. We experimentallyperform our algorithm on generated sparse signals and also correlated featurehistograms as multiview sparse sources from a multiview image database. Theresults show that RAMSIA significantly outperforms both classical CS and CSwith single SI, and RAMSIA with higher number of SIs gained more than the onewith smaller number of SIs.",
keywords = "Side information, math.OC, Compressive sensing, Sparse signal, n-`1 minimization, Adaptive weights",
author = "Luong, {Huynh Van} and J{\"u}rgen Seiler and Andr{\'e} Kaup and S{\o}ren Forchhammer",
year = "2016",
doi = "10.1109/ICIP.2016.7532816",
language = "English",
isbn = "9781467399616",
pages = "2534--2538",
booktitle = "Proceedings of IEEE International Conference on Image Processing 2016",
publisher = "IEEE",
address = "United States",

}

Luong, HV, Seiler, J, Kaup, A & Forchhammer, S 2016, Sparse Signal Reconstruction with Multiple Side Information using Adaptive Weights for Multiview Sources. in Proceedings of IEEE International Conference on Image Processing 2016. IEEE, pp. 2534-2538, 23rd IEEE International Conference on Image Processing , Phoenix, AZ, United States, 25/09/2016. https://doi.org/10.1109/ICIP.2016.7532816

Sparse Signal Reconstruction with Multiple Side Information using Adaptive Weights for Multiview Sources. / Luong, Huynh Van; Seiler, Jürgen; Kaup, André; Forchhammer, Søren.

Proceedings of IEEE International Conference on Image Processing 2016. IEEE, 2016. p. 2534-2538.

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

TY - GEN

T1 - Sparse Signal Reconstruction with Multiple Side Information using Adaptive Weights for Multiview Sources

AU - Luong, Huynh Van

AU - Seiler, Jürgen

AU - Kaup, André

AU - Forchhammer, Søren

PY - 2016

Y1 - 2016

N2 - This work considers reconstructing a target signal in a context ofdistributed sparse sources. We propose an efficient reconstruction algorithmwith the aid of other given sources as multiple side information (SI). Theproposed algorithm takes advantage of compressive sensing (CS) with SI andadaptive weights by solving a proposed weighted $n$-$\ell_{1}$ minimization.The proposed algorithm computes the adaptive weights in two levels, first eachindividual intra-SI and then inter-SI weights are iteratively updated at everyreconstructed iteration. This two-level optimization leads the proposedreconstruction algorithm with multiple SI using adaptive weights (RAMSIA) torobustly exploit the multiple SIs with different qualities. We experimentallyperform our algorithm on generated sparse signals and also correlated featurehistograms as multiview sparse sources from a multiview image database. Theresults show that RAMSIA significantly outperforms both classical CS and CSwith single SI, and RAMSIA with higher number of SIs gained more than the onewith smaller number of SIs.

AB - This work considers reconstructing a target signal in a context ofdistributed sparse sources. We propose an efficient reconstruction algorithmwith the aid of other given sources as multiple side information (SI). Theproposed algorithm takes advantage of compressive sensing (CS) with SI andadaptive weights by solving a proposed weighted $n$-$\ell_{1}$ minimization.The proposed algorithm computes the adaptive weights in two levels, first eachindividual intra-SI and then inter-SI weights are iteratively updated at everyreconstructed iteration. This two-level optimization leads the proposedreconstruction algorithm with multiple SI using adaptive weights (RAMSIA) torobustly exploit the multiple SIs with different qualities. We experimentallyperform our algorithm on generated sparse signals and also correlated featurehistograms as multiview sparse sources from a multiview image database. Theresults show that RAMSIA significantly outperforms both classical CS and CSwith single SI, and RAMSIA with higher number of SIs gained more than the onewith smaller number of SIs.

KW - Side information

KW - math.OC

KW - Compressive sensing

KW - Sparse signal

KW - n-`1 minimization

KW - Adaptive weights

U2 - 10.1109/ICIP.2016.7532816

DO - 10.1109/ICIP.2016.7532816

M3 - Article in proceedings

SN - 9781467399616

SP - 2534

EP - 2538

BT - Proceedings of IEEE International Conference on Image Processing 2016

PB - IEEE

ER -

Luong HV, Seiler J, Kaup A, Forchhammer S. Sparse Signal Reconstruction with Multiple Side Information using Adaptive Weights for Multiview Sources. In Proceedings of IEEE International Conference on Image Processing 2016. IEEE. 2016. p. 2534-2538 https://doi.org/10.1109/ICIP.2016.7532816