Distance Metric Learning for Content Identification
This paper considers a distance metric learning (DML) algorithm for a fingerprinting system, which identifies a query content by finding the fingerprint in the database (DB) that measures the shortest distance to the query fingerprint. For a given training set consisting of original and distorted fi...
Published in: | IEEE Transactions on Information Forensics and Security |
---|---|
Main Authors: | , , |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
2010
|
Subjects: | |
Online Access: | http://hdl.handle.net/10203/95203 http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=000284360000029&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=395d0a69a77a4892902e43d8987013d5 https://doi.org/10.1109/TIFS.2010.2064769 |
id |
ftkoasas:oai:koasas.kaist.ac.kr:10203/95203 |
---|---|
record_format |
openpolar |
spelling |
ftkoasas:oai:koasas.kaist.ac.kr:10203/95203 2023-05-15T16:01:32+02:00 Distance Metric Learning for Content Identification Jang, D Yoo, CD Yoo, Chang Dong Kalker, T 201012 http://hdl.handle.net/10203/95203 http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=000284360000029&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=395d0a69a77a4892902e43d8987013d5 https://doi.org/10.1109/TIFS.2010.2064769 ENG eng IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC ARTICLE A 2010 ftkoasas https://doi.org/10.1109/TIFS.2010.2064769 2013-12-15T18:30:08Z This paper considers a distance metric learning (DML) algorithm for a fingerprinting system, which identifies a query content by finding the fingerprint in the database (DB) that measures the shortest distance to the query fingerprint. For a given training set consisting of original and distorted fingerprints, a distance metric equivalent to the l(p) norm of the difference of two linearly projected fingerprints is learned by minimizing the false-positive rate (probability of perceptually dissimilar content to be identified as being similar) for a given false-negative rate (probability of perceptually similar content to be identified as being dissimilar). The learned metric can perform better than the often used l(p) distance and improve the robustness against a set of unexpected distortions. In the experiments, the distance metric learned by the proposed algorithm performed better than those metrics learned by well-known DML algorithms for classification. 전기및전자공학과 Article in Journal/Newspaper DML Korea Advanced Institute of Science and Technology: KOASAS - KAIST Open Access Self-Archiving System IEEE Transactions on Information Forensics and Security 5 4 932 944 |
institution |
Open Polar |
collection |
Korea Advanced Institute of Science and Technology: KOASAS - KAIST Open Access Self-Archiving System |
op_collection_id |
ftkoasas |
language |
English |
description |
This paper considers a distance metric learning (DML) algorithm for a fingerprinting system, which identifies a query content by finding the fingerprint in the database (DB) that measures the shortest distance to the query fingerprint. For a given training set consisting of original and distorted fingerprints, a distance metric equivalent to the l(p) norm of the difference of two linearly projected fingerprints is learned by minimizing the false-positive rate (probability of perceptually dissimilar content to be identified as being similar) for a given false-negative rate (probability of perceptually similar content to be identified as being dissimilar). The learned metric can perform better than the often used l(p) distance and improve the robustness against a set of unexpected distortions. In the experiments, the distance metric learned by the proposed algorithm performed better than those metrics learned by well-known DML algorithms for classification. 전기및전자공학과 |
format |
Article in Journal/Newspaper |
author |
Jang, D Yoo, CD Yoo, Chang Dong Kalker, T |
spellingShingle |
Jang, D Yoo, CD Yoo, Chang Dong Kalker, T Distance Metric Learning for Content Identification |
author_facet |
Jang, D Yoo, CD Yoo, Chang Dong Kalker, T |
author_sort |
Jang, D |
title |
Distance Metric Learning for Content Identification |
title_short |
Distance Metric Learning for Content Identification |
title_full |
Distance Metric Learning for Content Identification |
title_fullStr |
Distance Metric Learning for Content Identification |
title_full_unstemmed |
Distance Metric Learning for Content Identification |
title_sort |
distance metric learning for content identification |
publisher |
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
publishDate |
2010 |
url |
http://hdl.handle.net/10203/95203 http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=000284360000029&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=395d0a69a77a4892902e43d8987013d5 https://doi.org/10.1109/TIFS.2010.2064769 |
genre |
DML |
genre_facet |
DML |
op_doi |
https://doi.org/10.1109/TIFS.2010.2064769 |
container_title |
IEEE Transactions on Information Forensics and Security |
container_volume |
5 |
container_issue |
4 |
container_start_page |
932 |
op_container_end_page |
944 |
_version_ |
1766397352216625152 |