Presentation | 2018-05-18 Contour Extraction of Transparent Objects Using Fully Convolutional Networks Ryosuke Kubota, Kiyoshi Kogure, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In this paper, we propose two methods to extract contours of transparent objects from a grayscale image using fully convolutional networks. The first method separates the pixels on contours from the rest of the image. The second method introduce another task to separate the pixels in the regions around contours from the rest of image. The experimental results show that the first method is able to extract contours of transparent objects, whereas the second method has better performance than the first method. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | contour / fully convolutional network / transparent objects / multitask learning |
Paper # | SIP2018-10,IE2018-10,PRMU2018-10,MI2018-10 |
Date of Issue | 2018-05-10 (SIP, IE, PRMU, MI) |
Conference Information | |
Committee | PRMU / MI / IE / SIP |
---|---|
Conference Date | 2018/5/17(2days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | |
Chair | Shinichi Sato(NII) / Kensaku Mori(Nagoya Univ.) / Takayuki Hamamoto(Tokyo Univ. of Science) / Masahiro Okuda(Univ. of Kitakyushu) |
Vice Chair | Hironobu Fujiyoshi(Chubu Univ.) / Yoshihisa Ijiri(Omron) / Yoshiki Kawata(Tokushima Univ.) / Yuichi Kimura(Kinki Univ.) / Kazuya Kodama(NII) / Hideaki Kimata(NTT) / Shogo Muramatsu(Niigata Univ.) / Naoyuki Aikawa(TUS) |
Secretary | Hironobu Fujiyoshi(AIST) / Yoshihisa Ijiri(NAIST) / Yoshiki Kawata(Aichi Inst. of Tech.) / Yuichi Kimura(Nagoya Inst. of Tech.) / Kazuya Kodama(Nagoya Univ.) / Hideaki Kimata(KDDI Research) / Shogo Muramatsu(Chiba Inst. of Tech.) / Naoyuki Aikawa(Takushoku Univ.) |
Assistant | Masato Ishii(NEC) / Yusuke Sugano(Osaka Univ.) / Ryo Haraguchi(Univ. of Hyogo) / Yasushi Hirano(Yamaguchi Univ.) / Yasutaka Matsuo(NHK) / Kazuya Hayase(NTT) / Masayoshi Nakamoto(Hiroshima Univ.ひろ) |
Paper Information | |
Registration To | Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Medical Imaging / Technical Committee on Image Engineering / Technical Committee on Signal Processing |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Contour Extraction of Transparent Objects Using Fully Convolutional Networks |
Sub Title (in English) | |
Keyword(1) | contour |
Keyword(2) | fully convolutional network |
Keyword(3) | transparent objects |
Keyword(4) | multitask learning |
1st Author's Name | Ryosuke Kubota |
1st Author's Affiliation | Kanazawa Institute of Technology(KIT) |
2nd Author's Name | Kiyoshi Kogure |
2nd Author's Affiliation | Kanazawa Institute of Technology(KIT) |
Date | 2018-05-18 |
Paper # | SIP2018-10,IE2018-10,PRMU2018-10,MI2018-10 |
Volume (vol) | vol.118 |
Number (no) | SIP-33,IE-34,PRMU-35,MI-36 |
Page | pp.pp.41-46(SIP), pp.41-46(IE), pp.41-46(PRMU), pp.41-46(MI), |
#Pages | 6 |
Date of Issue | 2018-05-10 (SIP, IE, PRMU, MI) |