DocumentCode :
719431
Title :
Classification Using Residual Vector Quantization with Markov-Bayesian Structure
Author :
Khan, Syed Irteza Ali ; Anderson, David V. ; Barnes, Christopher F.
Author_Institution :
Nat. Univ. of Sci. & Technol., Islamabad, Pakistan
fYear :
2015
fDate :
7-9 April 2015
Firstpage :
454
Lastpage :
454
Abstract :
In this work, for a given a set of code vector assignments to an input by a multistage residual vector quantizer RVQ [1], Bayesian framework is formulated to find the most probable class membership of the input. Furthermore, Markov structure is also used to improve the memory cost of the classification.
Keywords :
Bayes methods; Markov processes; encoding; quantisation (signal); signal classification; Bayesian framework; Markov structure; Markov-Bayesian structure; code vector assignments; memory cost improvement; most probable class membership; multistage residual vector quantizer; residual vector quantization; signal classification; Accuracy; Bayes methods; Computers; Data compression; IEEE members; Markov processes; Training;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Data Compression Conference (DCC), 2015
Conference_Location :
Snowbird, UT
ISSN :
1068-0314
Type :
conf
DOI :
10.1109/DCC.2015.63
Filename :
7149317
Link To Document :
بازگشت