| Presentation | 2023-03-26 Analysis of learning performance in CycleGAN by applying data augmentation to few data Syuhei Kanzaki, Hidehiro Nakano, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In machine learning and deep learning, a huge amount of data is required for training. The image generation model GAN exists as a method to supplement the huge amount of training data. Data Augmentation is one of the methods to increase the number of data. It has been shown that the application of Data Augmentation to GANs can improve the performance of GANs. Data Augmentation was applied to CycleGAN, which is a combination of two GANs, for the model and training data, and the results showed the stability of learning with a small number of training data. In this study, we analyze the learning performance of the methods to add for training datasets based on data augmentation methods in CycleGAN. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Deep Learning / GAN / CycleGAN / Data Augmentation |
| Paper # | CCS2022-72 |
| Date of Issue | 2023-03-19 (CCS) |
| Conference Information | |
| Committee | CCS |
|---|---|
| Conference Date | 2023/3/26(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | RUSUTSU RESORT |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | Megumi Akai(Hokkaido Univ.) |
| Vice Chair | Hidehiro Nakano(Tokyo City Univ.) / Masaki Aida(TMU) |
| Secretary | Hidehiro Nakano(Shibaura Inst. of Tech.) / Masaki Aida(Mie Univ.) |
| Assistant | Hiroyuki Yasuda(Univ. of Tokyo) / Hiroyasu Ando(Tsukuba Univ.) / Tomoyuki Sasaki(Shonan Inst. of Tech.) / Miki Kobayashi(Rissho Univ.) |
| Paper Information | |
| Registration To | Technical Committee on Complex Communication Sciences |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Analysis of learning performance in CycleGAN by applying data augmentation to few data |
| Sub Title (in English) | |
| Keyword(1) | Deep Learning |
| Keyword(2) | GAN |
| Keyword(3) | CycleGAN |
| Keyword(4) | Data Augmentation |
| 1st Author's Name | Syuhei Kanzaki |
| 1st Author's Affiliation | Tokyo City University(Tokyo City Univ.) |
| 2nd Author's Name | Hidehiro Nakano |
| 2nd Author's Affiliation | Tokyo City University(Tokyo City Univ.) |
| Date | 2023-03-26 |
| Paper # | CCS2022-72 |
| Volume (vol) | vol.122 |
| Number (no) | CCS-453 |
| Page | pp.pp.54-58(CCS), |
| #Pages | 5 |
| Date of Issue | 2023-03-19 (CCS) |