Probabilistic combination of static and dynamic gait features for verification
Probabilistic combination of static and dynamic gait features for verification
This paper describes a novel probabilistic framework for biometric identification and data fusion. Based on intra and inter-class variation extracted from training data, posterior probabilities describing the similarity between two feature vectors may be directly calculated from the data using the logistic function and Bayes rule. Using a large publicly available database we show the two imbalanced gait modalities may be fused using this framework. All fusion methods tested provide an improvement over the best modality, with the weighted sum rule giving the best performance, hence showing that highly imbalanced classifiers may be fused in a probabilistic setting; improving not only the performance, but also generalized application capability.
Probabilistic, Biometrics, Gait, Bayesian, Logistic function, Fusion
23-30
Bazin, Alex I.
feead1f3-0fc6-4a1e-b089-f62361614633
Nixon, Mark S.
2b5b9804-5a81-462a-82e6-92ee5fa74e12
Jain, Anil K.
fdc5a2d3-c5f5-4124-b695-b718f004332f
Ratha, Nalini K.
ba80d3da-790c-478d-87c8-bb57ddb1489b
2005
Bazin, Alex I.
feead1f3-0fc6-4a1e-b089-f62361614633
Nixon, Mark S.
2b5b9804-5a81-462a-82e6-92ee5fa74e12
Jain, Anil K.
fdc5a2d3-c5f5-4124-b695-b718f004332f
Ratha, Nalini K.
ba80d3da-790c-478d-87c8-bb57ddb1489b
Bazin, Alex I. and Nixon, Mark S.
(2005)
Probabilistic combination of static and dynamic gait features for verification.
Jain, Anil K. and Ratha, Nalini K.
(eds.)
Biometric Technology for Human Identification II, SPIE Defense and Security Symposium, Orlando (Kissimmee), Florida, United States.
.
Record type:
Conference or Workshop Item
(Other)
Abstract
This paper describes a novel probabilistic framework for biometric identification and data fusion. Based on intra and inter-class variation extracted from training data, posterior probabilities describing the similarity between two feature vectors may be directly calculated from the data using the logistic function and Bayes rule. Using a large publicly available database we show the two imbalanced gait modalities may be fused using this framework. All fusion methods tested provide an improvement over the best modality, with the weighted sum rule giving the best performance, hence showing that highly imbalanced classifiers may be fused in a probabilistic setting; improving not only the performance, but also generalized application capability.
More information
Published date: 2005
Additional Information:
Event Dates: March 2005
Venue - Dates:
Biometric Technology for Human Identification II, SPIE Defense and Security Symposium, Orlando (Kissimmee), Florida, United States, 2005-03-01
Keywords:
Probabilistic, Biometrics, Gait, Bayesian, Logistic function, Fusion
Organisations:
Southampton Wireless Group
Identifiers
Local EPrints ID: 260722
URI: http://eprints.soton.ac.uk/id/eprint/260722
PURE UUID: 95950f5f-51c5-4989-983f-a1f6a43d7aea
Catalogue record
Date deposited: 07 Apr 2005
Last modified: 15 Mar 2024 02:35
Export record
Contributors
Author:
Alex I. Bazin
Editor:
Anil K. Jain
Editor:
Nalini K. Ratha
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
