| Presentation | 2020-03-02 Generating Adversarial Videos that Bypass Content Filtering Norihito Omori, Tatsuya Mori, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | A huge number of user-generated contents (UGC) are being uploaded on prominent internet video sharing sites such as YouTube. Such UGC includes many inappropriate content such as violence and discrimination. As a promising approach to automatically detect such inappropriate content from a huge volume of uploaded movies, machine-learning approaches using neural networks are increasingly becoming popular. On the other hand, it is well known that neural networks have an inherent vulnerability called ``adversarial input''. That is, it is possible to intentionally let a neural network misclassify an input by generating adversarial inputs. In this paper, we attempt to study whether an attacker can generate adversarial inputs against a neural network-based inappropriate content filtering system. This paper studies how to construct an effective adversarial input for a white box implementation that identifies whether or not an uploaded video contains violent scenes. We also perform user study that aims to study how generated adversarial input is perceived by human. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | video content filtering / adversarial input / neural network |
| Paper # | ICSS2019-95 |
| Date of Issue | 2020-02-24 (ICSS) |
| Conference Information | |
| Committee | ICSS / IPSJ-SPT |
|---|---|
| Conference Date | 2020/3/2(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Okinawa-Ken-Seinen-Kaikan |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Security, Trust, etc. |
| Chair | Hiroki Takakura(NII) |
| Vice Chair | Katsunari Yoshioka(Yokohama National Univ.) / Kazunori Kamiya(NTT) |
| Secretary | Katsunari Yoshioka(NICT) / Kazunori Kamiya(KDDI labs.) |
| Assistant | Keisuke Kito(Mitsubishi Electric) / Toshihiro Yamauchi(Okayama Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Information and Communication System Security / Special Interest Group on Security Psychology and Trust |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Generating Adversarial Videos that Bypass Content Filtering |
| Sub Title (in English) | |
| Keyword(1) | video content filtering |
| Keyword(2) | adversarial input |
| Keyword(3) | neural network |
| 1st Author's Name | Norihito Omori |
| 1st Author's Affiliation | Waseda University(Waseda Univ.) |
| 2nd Author's Name | Tatsuya Mori |
| 2nd Author's Affiliation | Waseda University(Waseda Univ.) |
| Date | 2020-03-02 |
| Paper # | ICSS2019-95 |
| Volume (vol) | vol.119 |
| Number (no) | ICSS-437 |
| Page | pp.pp.201-206(ICSS), |
| #Pages | 6 |
| Date of Issue | 2020-02-24 (ICSS) |