Presentation 2021-03-03
Analysis of deep convolutional neural network texture representation using Portilla-Simoncelli statistics
Yusuke Hamano, Hayaru Shouno,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Recently, DCNN has achieved significant success in the field of computer vision. It is suggested that the DCNN, which are even optimized to the visual recognition task, might obtain a varieties of texture representations. However, the property of the representations has not been revealed. In our work, To analyze the inner representation of VGG19, we introduced Portilla-Simoncelli Statistics which is a parametric texture model based on psychology. As the results, we suggested that VGG lacks the ability to represent features defined between different image scales of PSS.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) DCNN / Texture / Portilla-Simoncelli Statistics (PSS) / VGG
Paper # NC2020-48
Date of Issue 2021-02-24 (NC)

Conference Information
Committee NC / MBE
Conference Date 2021/3/3(3days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English) Neuro Computing, Medical Engineering, etc.
Chair Kazuyuki Samejima(Tamagawa Univ) / Takashi Watanabe(Tohoku Univ.)
Vice Chair Rieko Osu(Waseda Univ.) / Ryuhei Okuno(Setsunan Univ.)
Secretary Rieko Osu(NTT) / Ryuhei Okuno(ATR)
Assistant Ken Takiyama(TUAT) / Nobuhiko Wagatsuma(Toho Univ.) / Akihiro Karashima(Tohoku Inst. of Tech.) / Jun Akazawa(Meiji Univ. of Integrative Medicine)

Paper Information
Registration To Technical Committee on Neurocomputing / Technical Committee on ME and Bio Cybernetics
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Analysis of deep convolutional neural network texture representation using Portilla-Simoncelli statistics
Sub Title (in English)
Keyword(1) DCNN
Keyword(2) Texture
Keyword(3) Portilla-Simoncelli Statistics (PSS)
Keyword(4) VGG
1st Author's Name Yusuke Hamano
1st Author's Affiliation The University of Electro-Communications(UEC)
2nd Author's Name Hayaru Shouno
2nd Author's Affiliation The University of Electro-Communications(UEC)
Date 2021-03-03
Paper # NC2020-48
Volume (vol) vol.120
Number (no) NC-403
Page pp.pp.31-36(NC),
#Pages 6
Date of Issue 2021-02-24 (NC)