| Presentation | 2019-03-17 Investigation of Training Dataset based on 3DCG Simulation for Road Sign Detection using Deep Learning Ryuto Kato, Satoshi Nishiguchi, Yasuharu Mizutani, Wataru Hashimoto, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Deep learning is utilized in a field of image recognition for automated driving. However, it is costly of the time to collect real images and create high realistic 3DCG models in generating training dataset for road sign detector. In this paper, we propose a method for generating training dataset for road sign detection in consideration of changes about visual aspect of road sign in real space within the framework using light 3DCG models. We train a detector with 3DCG Image Dataset of Signs we proposed or Real Image Dataset of Signs augmented by geometric changing to a real image of sign. Furthermore, we compare these two datasets in the precision of detection about images for validation, and thereby discuss the effectiveness of training dataset we proposed. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Object Detection / YOLO / Deep Learning / Training Dataset / Data Augmentation / 3DCG |
| Paper # | BioX2018-29,PRMU2018-133 |
| Date of Issue | 2019-03-10 (BioX, PRMU) |
| Conference Information | |
| Committee | PRMU / BioX |
|---|---|
| Conference Date | 2019/3/17(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Shinichi Sato(NII) / Kazuhiko Sumi(AGU) |
| Vice Chair | Yoshihisa Ijiri(Omron) / Toru Tamaki(Hiroshima Univ.) / Hitoshi Imaoka(NEC) / Tetsushi Ohki(Shizuoka Univ.) |
| Secretary | Yoshihisa Ijiri(NEC) / Toru Tamaki(Osaka Univ.) / Hitoshi Imaoka(Fujitsu Labs.) / Tetsushi Ohki(Univ. of Electro-Comm.) |
| Assistant | Go Irie(NTT) / Yoshitaka Ushiku(Univ. of Tokyo) / Norihiro Okui(KDDI Research) / Daishi Watabe(Saitama Inst. of Tech.) |
| Paper Information | |
| Registration To | Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Biometrics |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Investigation of Training Dataset based on 3DCG Simulation for Road Sign Detection using Deep Learning |
| Sub Title (in English) | |
| Keyword(1) | Object Detection |
| Keyword(2) | YOLO |
| Keyword(3) | Deep Learning |
| Keyword(4) | Training Dataset |
| Keyword(5) | Data Augmentation |
| Keyword(6) | 3DCG |
| 1st Author's Name | Ryuto Kato |
| 1st Author's Affiliation | Osaka Institute of Technology(OIT) |
| 2nd Author's Name | Satoshi Nishiguchi |
| 2nd Author's Affiliation | Osaka Institute of Technology(OIT) |
| 3rd Author's Name | Yasuharu Mizutani |
| 3rd Author's Affiliation | Osaka Institute of Technology(OIT) |
| 4th Author's Name | Wataru Hashimoto |
| 4th Author's Affiliation | Osaka Institute of Technology(OIT) |
| Date | 2019-03-17 |
| Paper # | BioX2018-29,PRMU2018-133 |
| Volume (vol) | vol.118 |
| Number (no) | BioX-512,PRMU-513 |
| Page | pp.pp.1-6(BioX), pp.1-6(PRMU), |
| #Pages | 6 |
| Date of Issue | 2019-03-10 (BioX, PRMU) |