| Presentation | 2021-11-25 Acceleration of Neural Network Visualization RISE by Reduction of Masks Tomoki Nakada, Imamura Kousuke, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Recently, various techniques have been proposed to visualize the basis for decisions in image recognition using neural networks. In RISE which is one method of them, the importance in an image is heat-mapped by observing the response of the network while partially hiding the input image with a randomly generated mask. However, due to the random mask generation, this method requires many masked images to obtain stationarity, which in turn requires a large amount of computation time. In this study, in addition to the random masks used in the conventional method, we use non-random patch masks that only pass through one limited area to speed up RISE by reducing the number of masks used. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Image recognition / Neural network visualization / RISE |
| Paper # | CS2021-61,IE2021-20 |
| Date of Issue | 2021-11-18 (CS, IE) |
| Conference Information | |
| Committee | IPSJ-AVM / CS / IE / ITE-BCT |
|---|---|
| Conference Date | 2021/11/25(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Online |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Image coding, Communications and streaming technologies, etc. |
| Chair | 笠井 裕之(早大) / Jun Terada(NTT) / Kazuya Kodama(NII) / Kyoichi Saito(NHK) |
| Vice Chair | / Daisuke Umehara(Kyoto Inst. of Tech.) / Hiroyuki Bandoh(NTT) / Toshihiko Yamazaki(Univ. of Tokyo) / Hidekazu Murata(Kyoto Univ.) / Hajime Saito(TV Tokyo) |
| Secretary | (早大) / Daisuke Umehara(KDDI) / Hiroyuki Bandoh(NTT) / Toshihiko Yamazaki(NICT) / Hidekazu Murata(NTT) / Hajime Saito(KDDI Research) |
| Assistant | / Takahiro Yamaura(Toshiba) / Yuta Ida(Yamaguchi Univ.) / Shunsuke Iwamura(NHK) / Shinobu Kudo(NTT) / Tatsuhiko Itokawa(Mitsubishi Electric) / Takumi Matsumoto(Furukawa Electric) / Takeshi Maruyama(Furukawa Electric) / Yoshie Enoki(TBS) |
| Paper Information | |
| Registration To | The Special Interest Groups of Audio Visual and Multimedia Information Processing / Technical Committee on Communication Systems / Technical Committee on Image Engineering / 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) | Acceleration of Neural Network Visualization RISE by Reduction of Masks |
| Sub Title (in English) | |
| Keyword(1) | Image recognition |
| Keyword(2) | Neural network visualization |
| Keyword(3) | RISE |
| 1st Author's Name | Tomoki Nakada |
| 1st Author's Affiliation | Kanazawa University(Kanazawa Univ.) |
| 2nd Author's Name | Imamura Kousuke |
| 2nd Author's Affiliation | Kanazawa University(Kanazawa Univ.) |
| Date | 2021-11-25 |
| Paper # | CS2021-61,IE2021-20 |
| Volume (vol) | vol.121 |
| Number (no) | CS-268,IE-269 |
| Page | pp.pp.7-12(CS), pp.7-12(IE), |
| #Pages | 6 |
| Date of Issue | 2021-11-18 (CS, IE) |