Presentation | 2019-03-17 Arbitrary Charactor Image Generation in Arbitrary Poses using Neural Network Hidemoto Nakada, Hideki Asoh, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Thanks to recent improvement of image generation technologies by neural networks, now we can gener- ate photo-realistic images. To control the content of the generated images, such as the person detail in the images or view angle of the images, is still an open problem. In this paper we introduce our effort to render a person image in arbitrary pose using few images of the person in different poses. The proposed system encodes the context images of the person into embeddings, generate target image embedding with them and the query pose embedding, and decode the embedding into the target image. We evaluated the system using synthesized data set generated from 3D person models. We found that the system produces images with reasonable quality for training data, while the quality for the test data is not satisfactory. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Image Generation / Attention Network / Deep Neural Network |
Paper # | BioX2018-41,PRMU2018-145 |
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) | Arbitrary Charactor Image Generation in Arbitrary Poses using Neural Network |
Sub Title (in English) | |
Keyword(1) | Image Generation |
Keyword(2) | Attention Network |
Keyword(3) | Deep Neural Network |
1st Author's Name | Hidemoto Nakada |
1st Author's Affiliation | National Institute of Advanced Science and Technology(AIST) |
2nd Author's Name | Hideki Asoh |
2nd Author's Affiliation | National Institute of Advanced Science and Technology(AIST) |
Date | 2019-03-17 |
Paper # | BioX2018-41,PRMU2018-145 |
Volume (vol) | vol.118 |
Number (no) | BioX-512,PRMU-513 |
Page | pp.pp.73-78(BioX), pp.73-78(PRMU), |
#Pages | 6 |
Date of Issue | 2019-03-10 (BioX, PRMU) |