Presentation 2020-10-01
Robustness Evaluation of Detectinon methods for Image manipulation with GANs
Miki Tanaka, Hitoshi Kiya,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Recent rapid advances in image manipulation tools and deep image synthesis techniques, such as Generative Adversarial Networks (GANs) have easily generated fake images, so detecting manipulated images has become an urgent issue. In this paper, we evaluate state-of-the-art methods for detecting images manipulated by using GANs in terms of robustness against various manipulation operations. Most forgery detection methods assume that images are generated by using a specific manipulation technique, and the methods aim to detect unique features caused by the manipulation technique. However, the detection performance has not been discussed sufficiently when a number of manipulation techniques such as JPEG compression are applied as the same time. In this paper, robustness of forgery detection methods is evaluated under the use of multiple manipulation techniques. In addition, we investigate a possibility that there is a method with robust hashing which is robust even when multiple manipulation techniques are carried out.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) fake images / GAN / robust / image manipulation
Paper # SIS2020-14
Date of Issue 2020-09-24 (SIS)

Conference Information
Committee SIS / ITE-BCT
Conference Date 2020/10/1(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English) System Implementation Technology, Short Range Wireless Systems, Smart Multimedia Systems, Broadcasting Technology, etc.
Chair Noriaki Suetake(Yamaguchi Univ.) / Kyoichi Saito(NHK)
Vice Chair Tomoaki Kimura(Kanagawa Inst. of Tech.) / Naoto Sasaoka(Tottori Univ.)
Secretary Tomoaki Kimura(Kindai Univ.) / Naoto Sasaoka(National Inst. of Tech., Ube College) / (NHK)
Assistant Yukihiro Bandoh(NTT) / Soh Yoshida(Kansai Univ.) / Shigeki Shiokawa(Kanagawa Inst. of Tech.) / Shoichiro Sekiguchi(NHK) / Toshiharu Morizumi(NTT) / Toshimitsu Kobayashi(NBN)

Paper Information
Registration To Technical Committee on Smart Info-Media Systems / Technical Group on Broadcasting and Communication Technologies
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Robustness Evaluation of Detectinon methods for Image manipulation with GANs
Sub Title (in English)
Keyword(1) fake images
Keyword(2) GAN
Keyword(3) robust
Keyword(4) image manipulation
1st Author's Name Miki Tanaka
1st Author's Affiliation Tokyo Metropolitan University(Tokyo Metropolitan Univ.)
2nd Author's Name Hitoshi Kiya
2nd Author's Affiliation Tokyo Metropolitan University(Tokyo Metropolitan Univ.)
Date 2020-10-01
Paper # SIS2020-14
Volume (vol) vol.120
Number (no) SIS-176
Page pp.pp.23-28(SIS),
#Pages 6
Date of Issue 2020-09-24 (SIS)