Presentation | 2016-02-22 Iteratively applying image inpainting with automatic detection of failure regions using a convolutional neural network Takahiro Tanaka, Norihiko Kawai, Yuta Nakashima, Tomokazu Sato, Naokazu Yokoya, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | Image inpainting is a method that fills in missing regions with plausible textures in an image to remove undesired objects for enhancing the utility of the image. Image inpainting has been widely studied; some conventional methods define an energy function based on similarity between missing regions and the rest of the image. They search for patterns that are similar to the missing regions and update pixel values so that the energy function is minimized. However, applying such methods only once does not always generate plausible textures for the missing regions. This problem can be solved by iteratively applying image inpainting with manually reselecting such failure regions. To improve the quality of image inpainting result without such manual reselection, this study proposes a method that iteratively applies image inpainting with automatic detection of failure regions in a resulting image using a convolutional neural network. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | image inpainting / convolutional neural network / automatic detection of failure regions |
Paper # | PRMU2015-160,CNR2015-61 |
Date of Issue | 2016-02-14 (PRMU, CNR) |
Conference Information | |
Committee | PRMU / CNR |
---|---|
Conference Date | 2016/2/21(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Eisaku Maeda(NTT) / Michita Imai(Keio Univ.) |
Vice Chair | Shuji Senda(NEC) / Seiichi Uchida(Kyushu Univ.) / Takahiro Miyashita(ATR) / Masayuki Kanbara(NAIST) |
Secretary | Shuji Senda(Mie Univ.) / Seiichi Uchida(DENSO IT Lab.) / Takahiro Miyashita(Univ. of Tokyo) / Masayuki Kanbara(Univ. of Tsukuba) |
Assistant | Kazuaki Kondo(Kyoto Univ.) / Akisato Kimura(NTT) / Kosuke Yoshioka(Panasonic) / Yuka Kobayashi(Toshiba) / Takahiro Matsumoto(NTT) |
Paper Information | |
Registration To | Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Cloud Network Robotics |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Iteratively applying image inpainting with automatic detection of failure regions using a convolutional neural network |
Sub Title (in English) | |
Keyword(1) | image inpainting |
Keyword(2) | convolutional neural network |
Keyword(3) | automatic detection of failure regions |
1st Author's Name | Takahiro Tanaka |
1st Author's Affiliation | Nara Institute of Science and Technology(NAIST) |
2nd Author's Name | Norihiko Kawai |
2nd Author's Affiliation | Nara Institute of Science and Technology(NAIST) |
3rd Author's Name | Yuta Nakashima |
3rd Author's Affiliation | Nara Institute of Science and Technology(NAIST) |
4th Author's Name | Tomokazu Sato |
4th Author's Affiliation | Nara Institute of Science and Technology(NAIST) |
5th Author's Name | Naokazu Yokoya |
5th Author's Affiliation | Nara Institute of Science and Technology(NAIST) |
Date | 2016-02-22 |
Paper # | PRMU2015-160,CNR2015-61 |
Volume (vol) | vol.115 |
Number (no) | PRMU-456,CNR-457 |
Page | pp.pp.133-138(PRMU), pp.133-138(CNR), |
#Pages | 6 |
Date of Issue | 2016-02-14 (PRMU, CNR) |