| Presentation | 2021-05-20 A universal detector of CNN-generated images based on properties of checkerboard artifacts Miki Tanaka, Hitoshi Kiya, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | We propose a universal detector of images generated by using any CNNs to detect CNN-generated images. We consider properties of checkerboard artifacts in the frequency domain that some CNN-generated images have, and the spectrum of images is emphasized in accordance with the properties. Next, emphasized spectrums are inputted into a detector to judges whether inputted images are CNN-generated ones or not. The proposed method is demonstrated to outperform a state-of-the-art method under some conditions. In addition, the combination of the proposed method and the conventional one is confirmed to improve the performance of these methods in an experiment. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | GAN / CNN / checkerboard artifact / fake image |
| Paper # | IT2021-3,EMM2021-3 |
| Date of Issue | 2021-05-13 (IT, EMM) |
| Conference Information | |
| Committee | EMM / IT |
|---|---|
| Conference Date | 2021/5/20(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Information Security, Information Theory, Information Hiding, etc. |
| Chair | Masaki Kawamura(Yamaguchi Univ.) / Tadashi Wadayama(Nagoya Inst. of Tech.) |
| Vice Chair | Motoi Iwata(Osaka Prefecture Univ.) / Masaaki Fujiyoshi(Tokyo Metropolitan Univ.) / Tetsuya Kojima(Tokyo Kosen) |
| Secretary | Motoi Iwata(Tokyo Denki Univ.) / Masaaki Fujiyoshi(Kansai Univ.) / Tetsuya Kojima(Yamaguchi Univ.) |
| Assistant | Madoka Hasegawa(Utsunomiya Univ.) / Maki Yoshida(NICT) / Takahiro Ohta(Senshu Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Enriched MultiMedia / Technical Committee on Information Theory |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | A universal detector of CNN-generated images based on properties of checkerboard artifacts |
| Sub Title (in English) | |
| Keyword(1) | GAN |
| Keyword(2) | CNN |
| Keyword(3) | checkerboard artifact |
| Keyword(4) | fake image |
| 1st Author's Name | Miki Tanaka |
| 1st Author's Affiliation | Tokyo metropolitan University(Metro Univ.) |
| 2nd Author's Name | Hitoshi Kiya |
| 2nd Author's Affiliation | Tokyo metropolitan University(Metro Univ.) |
| Date | 2021-05-20 |
| Paper # | IT2021-3,EMM2021-3 |
| Volume (vol) | vol.121 |
| Number (no) | IT-28,EMM-29 |
| Page | pp.pp.13-18(IT), pp.13-18(EMM), |
| #Pages | 6 |
| Date of Issue | 2021-05-13 (IT, EMM) |