Presentation 2017-09-04
Privacy-preserving SVM processing by using random unitary transformation
Takahiro Maekawa, Yuma Kinoshita, Sayaka Shiota, Hitoshi Kiya,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper,we propose a privacy-preserving SVM processing method with templates protected by using a random unitary transformation, and evaluate the effectiveness of the proposed scheme, where templates mean features extracted from images or biometric traits. Recently, cloud computing is spreading in many fields. However, the cloud computing has some serious issues for end users, such as unauthorized use and leak of data, and privacy compromise, due to unreliability of providers and some accident. Because of such a situation, this paper considers privacy-preserving SVM processing. It is shown that the unitary transformation-based template protection method enables us to not only protect templates, but also have the same classification performance as that of SVM processing with unprotected templates. Moreover, we consider how to update/assign secret keys, and then describe that it is possible to provide a higher performance than with unprotected templates. Some face recognition experiments are carried out as a SVM classification scheme to experimentally confirm the effectiveness of the method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) unitary transform / kernel method / SVM / template protection method / biometrics
Paper # LOIS2017-15,IE2017-36,EMM2017-44
Date of Issue 2017-08-28 (LOIS, IE, EMM)

Conference Information
Committee EMM / IE / LOIS / IEE-CMN / ITE-ME
Conference Date 2017/9/4(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Kyoto Univ. (Clock Tower Centennial Hall)
Topics (in Japanese) (See Japanese page)
Topics (in English) Multimedia Communication/System, Lifelog Applications, IP Broadcasting/Video Transmission, Media Security, etc.
Chair Keiichi Iwamura(TUS) / Takayuki Hamamoto(Tokyo Univ. of Science) / Hiroyuki Nishi(Sojo Univ.) / Yoshimi Fukumura(Nagaoka Univ. of Tech.) / Miki Haseyama(Hokkaido Univ.)
Vice Chair Hirohisa Hioki(Kyoto Univ.) / Minoru Kuribayashi(Okayama Univ.) / Kazuya Kodama(NII) / Hideaki Kimata(NTT) / Tomohiro Yamada(NTT) / / Norio Tagawa(Tokyo Metropolitan Univ.)
Secretary Hirohisa Hioki(Shizuoka Univ.) / Minoru Kuribayashi(Tokyo Metropolitan Univ.) / Kazuya Kodama(Nagoya Univ.) / Hideaki Kimata(KDDI Research) / Tomohiro Yamada(Nagasaki Univ.) / (NTT) / Norio Tagawa(CRIEPI)
Assistant Kan Hyonho(NIT, Tokyo) / Harumi Murata(Chukyo Univ.) / Yasutaka Matsuo(NHK) / Kazuya Hayase(NTT) / Motoi Okamoto(Research Organization of Information and Systems) / Keisuke Utsu(Tokai Univ.) / Takeshi YAGI(NTT)

Paper Information
Registration To Technical Committee on Enriched MultiMedia / Technical Committee on Image Engineering / Technical Committee on Life Intelligence and Office Information Systems / Technical Meeting on Communications / Technical Group on Media Engineering
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Privacy-preserving SVM processing by using random unitary transformation
Sub Title (in English)
Keyword(1) unitary transform
Keyword(2) kernel method
Keyword(3) SVM
Keyword(4) template protection method
Keyword(5) biometrics
1st Author's Name Takahiro Maekawa
1st Author's Affiliation Tokyo Metropolitan University(Tokyo Metro.Univ.)
2nd Author's Name Yuma Kinoshita
2nd Author's Affiliation Tokyo Metropolitan University(Tokyo Metro.Univ.)
3rd Author's Name Sayaka Shiota
3rd Author's Affiliation Tokyo Metropolitan University(Tokyo Metro.Univ.)
4th Author's Name Hitoshi Kiya
4th Author's Affiliation Tokyo Metropolitan University(Tokyo Metro.Univ.)
Date 2017-09-04
Paper # LOIS2017-15,IE2017-36,EMM2017-44
Volume (vol) vol.117
Number (no) LOIS-199,IE-200,EMM-201
Page pp.pp.13-18(LOIS), pp.13-18(IE), pp.13-18(EMM),
#Pages 6
Date of Issue 2017-08-28 (LOIS, IE, EMM)