A parametric copula based framework for multimodal signal processing

We present a framework for the joint processing of multimodal data such as audio-video data streams. We first consider the problem of estimating the joint distribution of statistically dependent multimodal random variables. We discuss the issues involved and provide a copula based solution. Applicat...

Full description

Saved in:
Bibliographic Details
Main Authors: Iyengar, S.G., Varshney, P.K., Damarla, T.
Format: Conference Proceeding
Language:English
Subjects:
Acoustic sensors
Cameras
Copula theory
Hypothesis testing
Kullback-Leibler distance
Multisensor fusion
Parametric statistics
Probability distribution
Random variables
Signal processing
Statistical dependence
Statistical distributions
Streaming media
Video sharing
Zinc
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We present a framework for the joint processing of multimodal data such as audio-video data streams. We first consider the problem of estimating the joint distribution of statistically dependent multimodal random variables. We discuss the issues involved and provide a copula based solution. Application of this approach to solve a multisensor fusion problem for the detection of a random event is also discussed.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2009.4959978