Design and analysis of entropy-constrained reflected residual vector quantization

Residual vector quantization (RVQ) is a vector quantization (VQ) paradigm which imposes structural constraints on the encoder in order to reduce the encoding search burden and memory storage requirements of an unconstrained VQ. Jointly optimized RVQ (JORVQ) is an effective design algorithm for minim...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Mousa, W.A.H. (author)
مؤلفون آخرون: Khan, M.A.U. (author), unknown (author)
التنسيق: article
منشور في: 2002
الموضوعات:
الوصول للمادة أونلاين:https://eprints.kfupm.edu.sa/id/eprint/14834/1/14834_1.pdf
https://eprints.kfupm.edu.sa/id/eprint/14834/2/14834_2.doc
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513394310119424
author Mousa, W.A.H.
author2 Khan, M.A.U.
unknown
author2_role author
author
author_facet Mousa, W.A.H.
Khan, M.A.U.
unknown
author_role author
dc.creator.none.fl_str_mv Mousa, W.A.H.
Khan, M.A.U.
unknown
dc.date.none.fl_str_mv 2002
2020
dc.format.none.fl_str_mv application/pdf
application/msword
dc.identifier.none.fl_str_mv https://eprints.kfupm.edu.sa/id/eprint/14834/1/14834_1.pdf
https://eprints.kfupm.edu.sa/id/eprint/14834/2/14834_2.doc
(2002) Design and analysis of entropy-constrained reflected residual vector quantization. Acoustics, Speech, and Signal Processing, 2002. Proceedings. (ICASSP '02). IEEE International conference, 3.
dc.language.none.fl_str_mv en
en
dc.publisher.none.fl_str_mv IEEE
dc.relation.none.fl_str_mv https://eprints.kfupm.edu.sa/id/eprint/14834/
dc.rights.*.fl_str_mv info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Computer
dc.title.none.fl_str_mv Design and analysis of entropy-constrained reflected residual vector quantization
dc.type.none.fl_str_mv Article
PeerReviewed
info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/article
description Residual vector quantization (RVQ) is a vector quantization (VQ) paradigm which imposes structural constraints on the encoder in order to reduce the encoding search burden and memory storage requirements of an unconstrained VQ. Jointly optimized RVQ (JORVQ) is an effective design algorithm for minimizing the overall quantization error. Reflected residual vector quantization (RRVQ) is an alternative design algorithm for the RVQ structure with a smaller computation burden. RRVQ works by imposing an additional symmetry constraint on the RVQ codebook design. Savings in computation were accompanied by an increase in distortion. However, an RRVQ codebook, being structured in nature, is expected to provide lower output entropy. Therefore, we generalize RRVQ to include noiseless entropy coding. The method is referred to as entropy-constrained RRVQ (EC-RRVQ). Simulation results show that EC-RRVQ outperforms RRVQ by 4 dB for memoryless Gaussian and Laplacian sources. In addition, for the same synthetic sources, EC-RRVQ provided an improvement over other entropy-constrained designs, such as entropy-constrained JORVQ (EC-JORVQ). The design performed equally well on image data. In comparison with EC-JORVQ, EC-RRVQ is simpler and outperforms the EC-JORVQ.
eu_rights_str_mv openAccess
format article
id KFUPM_df8ff9b961e80ee0f92152bca241e13b
identifier_str_mv (2002) Design and analysis of entropy-constrained reflected residual vector quantization. Acoustics, Speech, and Signal Processing, 2002. Proceedings. (ICASSP '02). IEEE International conference, 3.
language_invalid_str_mv en
network_acronym_str KFUPM
network_name_str King Fahd University of Petroleum and Minerals
oai_identifier_str oai::14834
publishDate 2002
publisher.none.fl_str_mv IEEE
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Design and analysis of entropy-constrained reflected residual vector quantizationMousa, W.A.H.Khan, M.A.U.unknownComputerResidual vector quantization (RVQ) is a vector quantization (VQ) paradigm which imposes structural constraints on the encoder in order to reduce the encoding search burden and memory storage requirements of an unconstrained VQ. Jointly optimized RVQ (JORVQ) is an effective design algorithm for minimizing the overall quantization error. Reflected residual vector quantization (RRVQ) is an alternative design algorithm for the RVQ structure with a smaller computation burden. RRVQ works by imposing an additional symmetry constraint on the RVQ codebook design. Savings in computation were accompanied by an increase in distortion. However, an RRVQ codebook, being structured in nature, is expected to provide lower output entropy. Therefore, we generalize RRVQ to include noiseless entropy coding. The method is referred to as entropy-constrained RRVQ (EC-RRVQ). Simulation results show that EC-RRVQ outperforms RRVQ by 4 dB for memoryless Gaussian and Laplacian sources. In addition, for the same synthetic sources, EC-RRVQ provided an improvement over other entropy-constrained designs, such as entropy-constrained JORVQ (EC-JORVQ). The design performed equally well on image data. In comparison with EC-JORVQ, EC-RRVQ is simpler and outperforms the EC-JORVQ.IEEE20022020ArticlePeerReviewedinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/articleapplication/pdfapplication/mswordhttps://eprints.kfupm.edu.sa/id/eprint/14834/1/14834_1.pdfhttps://eprints.kfupm.edu.sa/id/eprint/14834/2/14834_2.doc (2002) Design and analysis of entropy-constrained reflected residual vector quantization. Acoustics, Speech, and Signal Processing, 2002. Proceedings. (ICASSP '02). IEEE International conference, 3. enenhttps://eprints.kfupm.edu.sa/id/eprint/14834/info:eu-repo/semantics/openAccessoai::148342019-11-01T14:07:40Z
spellingShingle Design and analysis of entropy-constrained reflected residual vector quantization
Mousa, W.A.H.
Computer
status_str publishedVersion
title Design and analysis of entropy-constrained reflected residual vector quantization
title_full Design and analysis of entropy-constrained reflected residual vector quantization
title_fullStr Design and analysis of entropy-constrained reflected residual vector quantization
title_full_unstemmed Design and analysis of entropy-constrained reflected residual vector quantization
title_short Design and analysis of entropy-constrained reflected residual vector quantization
title_sort Design and analysis of entropy-constrained reflected residual vector quantization
topic Computer
url https://eprints.kfupm.edu.sa/id/eprint/14834/1/14834_1.pdf
https://eprints.kfupm.edu.sa/id/eprint/14834/2/14834_2.doc