Residual Vector Quantizers with Jointly Optimized Code Books

Barnes, Christopher F.; Frost, R.L.

doi:10.1016/s0065-2539(08)61034-3

Cited by 22 publications

(2 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since it is usually impossible to simultaneously optimize the ratedistortion performance in all stages, a weighted average of the distortion at each stage of refinement has been minimized as a design criterion in [10]. A different approach to solve a related problem can be found in [11].…”

Section: Introductionmentioning

confidence: 99%

Design of successively refinable trellis-coded quantizers

Jafarkhani

Tarokin

1999

IEEE Trans. Inform. Theory

View full text Add to dashboard Cite

We propose successively refinable trellis-coded quantizers for progressive transmission. (Progressive transmission is an essential component of image and multimedia browsing systems.) A new trellis structure which is scalable is used in the design of our trellis-coded quantizers. A hierarchical set partitioning is developed to preserve successive refinability. Two algorithms for designing trellis-coded quantizers which provide embedded bit streams are provided. The computational complexity of the proposed schemes is compared with that of trellis-coded quantization. Simulation results show excellent performances for memoryless sources.

show abstract

Section: Introductionmentioning

confidence: 99%

Design of successively refinable trellis-coded quantizers

Jafarkhani

Tarokin

1999

IEEE Trans. Inform. Theory

View full text Add to dashboard Cite

show abstract

“…At the expense of increased computations, this performance improvement grows to 10.77 dB for the M = 8 case. Although not all details of this design process are given here, some details can be found in[6]. Use of similar design algorithms has been reported in[7, 8, 91.…”

mentioning

confidence: 99%

Classified variable rate residual vector quantization applied to image subband coding

Barnes

Holder

[Proceedings] DCC `93: Data Compression Conference

View full text Add to dashboard Cite

Multi-stage residual vector quantization (RVQ) is applied to image subband coding. The linear growth with the dimension-rate product of RVQ computation and memory requirements permits practical implementations of RVQs with large dimensions or high rates. This feature is exploited in subband coding by quantizing low-resolution subbands with small-dimension high-rate RVQs, and high-resolution subbands with largedimension low-rate RVQs. The RVQ vector sizes vary by a factor of four in parallel with the decimation and upsampling processes from one resolution level to the next. Two forms of rate allocation are achieved with the RVQ subband system. A type of concentric shell partitioned vector classifier with side information is used to separate noise-like subband vectors from structured subband vectors. For the largedimension low-rate RVQs, variable rate RVQ with side information is used to permit different numbers of RVQ stages to be used on different vectors within a concentric shell partition class.

show abstract