| Presentation | 2018-09-27 [Special Talk] Coded Acquistion of Light Fields Keita Takahashi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | A light field, which is often understood as a set of dense multi-view images, has been utilized in various 2D/3D applications. Acquiring a light field is a challenging task due to the amount of data. To make the acquisition process efficient, coded aperture cameras were successfully adopted; using these cameras, a light field is computationally reconstructed from several images that are acquired with different aperture patterns. However, it is still difficult to reconstruct a high-quality light field from only a few acquired images. Previously, this problem has often been discussed from the context of compressed sensing (CS), where sparse representations on a pre-trained dictionary or basis are explored to reconstruct the light field. We first took an approach to this problem from the perspective of principal component analysis (PCA) and non-negative matrix factorization (NMF), where only a small number of basis vectors are selected in advance based on the analysis of the training dataset. We also proposed a learning-based framework, where the entire pipeline of light field acquisition was formulated from the perspective of an auto-encoder that can be trained end-to-end by using a collection of training samples. We obtained promising results both with simulative experiments and a real camera prototype. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Light field / Coded aperture / PCA / NMF / Deep learing / Compressed sensing |
| Paper # | LOIS2018-15,IE2018-35,EMM2018-54 |
| Date of Issue | 2018-09-20 (LOIS, IE, EMM) |
| Conference Information | |
| Committee | IEE-CMN / EMM / LOIS / IE / ITE-ME |
|---|---|
| Conference Date | 2018/9/27(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Beppu Int'l Convention Ctr. aka B-CON Plaza |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | Multimedia Communication/System, Lifelog Applications, IP Broadcasting/Video Transmission, Media Security, Media Processing (AI, Deep Learning), etc. |
| Chair | Shun Morimura(CRIEPI) / Keiichi Iwamura(TUC) / Tomohiro Yamada(NTT) / Takayuki Hamamoto(Tokyo Univ. of Science) / Miki Haseyama(北大) |
| Vice Chair | / Minoru Kuribayashi(Okayama Univ.) / Tetsuya Kojima(NIT,Tokyo College) / Toru Kobayashi(Nagasaki Univ.) / Hideaki Kimata(NTT) / Kazuya Kodama(NII) / Norio Tagawa(Tokyo Metropolitan Univ.) |
| Secretary | (Tokai Univ.) / Minoru Kuribayashi(Kansai Univ.) / Tetsuya Kojima(NIT, Tokyo) / Toru Kobayashi(Chukyo Univ.) / Hideaki Kimata(NTT) / Kazuya Kodama(Research Organization of Information and Systems) / Norio Tagawa(KDDI Research) |
| Assistant | Tomotaka Kimura(Doshisha Univ.) / 田中 彰浩(CRIEPI) / Hiroko Akiyama(NIT, Nagano College) / Kitahiro Kaneda(CANON) / Shinichiro Eitoku(NTT) / Kazuya Hayase(NTT) / Yasutaka Matsuo(NHK) |
| Paper Information | |
| Registration To | Technical Meeting on Communications / Technical Committee on Enriched MultiMedia / Technical Committee on Life Intelligence and Office Information Systems / Technical Committee on Image Engineering / Technical Group on Media Engineering |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | [Special Talk] Coded Acquistion of Light Fields |
| Sub Title (in English) | From Basis Representation to Deep Learning |
| Keyword(1) | Light field |
| Keyword(2) | Coded aperture |
| Keyword(3) | PCA |
| Keyword(4) | NMF |
| Keyword(5) | Deep learing |
| Keyword(6) | Compressed sensing |
| 1st Author's Name | Keita Takahashi |
| 1st Author's Affiliation | Nagoya University(Nagoya Univ.) |
| Date | 2018-09-27 |
| Paper # | LOIS2018-15,IE2018-35,EMM2018-54 |
| Volume (vol) | vol.118 |
| Number (no) | LOIS-222,IE-223,EMM-224 |
| Page | pp.pp.29-30(LOIS), pp.29-30(IE), pp.29-30(EMM), |
| #Pages | 2 |
| Date of Issue | 2018-09-20 (LOIS, IE, EMM) |