Presentation | 2018-05-25 Optimal Design and Coded Image Quality Assessment of the Multi-view and Super-resolution Images Based on Structure of Convolutional Neural Network Norifumi Kawabata, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | The image screen resolution by viewpoints is low, comparing to single-view images since there are many viewpoints for multi-view images. Therefore, the super-resolution image processing is often carried out in the case of presenting image. In the case of transforming from low to high resolution image, the number of output data is more than that of input data. From this, there are many studies for super-resolution processing method using neural network. Furthermore, we can come to approach on super-resolution processing based on deep learning by appearing deep learning tools. These performance are shown by applying the only deep learning theory for super-resolution processing. However, we consider that the optimal condition and design for super-resolution processing are achieved better by improving these algorithms and setting parameter appropriately. In this paper, first, we carried out experiments on optimal condition and design of super-resolution processing for the multi-view 3D images encoded and decoded by H.265/HEVC, focused on structure of convolutional neural network by using Chainer. And then, we assessed for the generated images quality objectively, and compare to each image. Finally, we discussed for experimental results. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | Multi-view 3D Image / Super-resolution / Chainer / Convolutional Neural Network (CNN) / H.265/HEVC / Peak Signal to Noise Ratio (PSNR) |
Paper # | IMQ2018-3 |
Date of Issue | 2018-05-18 (IMQ) |
Conference Information | |
Committee | IMQ |
---|---|
Conference Date | 2018/5/25(1days) |
Place (in Japanese) | (See Japanese page) |
Place (in English) | Chibe Institute of Technology, Tsudanuma Campus |
Topics (in Japanese) | (See Japanese page) |
Topics (in English) | Image Media Quality |
Chair | Kenji Sugiyama(Seikei Univ.) |
Vice Chair | Toshiya Nakaguchi(Chiba Univ.) / Mitsuru Maeda(Canon) |
Secretary | Toshiya Nakaguchi(Nagoya Univ.) / Mitsuru Maeda(Sony) |
Assistant | Masaru Tsuchida(NTT) / Gosuke Ohashi(Shizuoka Univ.) |
Paper Information | |
Registration To | Technical Committee on Image Media Quality |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Optimal Design and Coded Image Quality Assessment of the Multi-view and Super-resolution Images Based on Structure of Convolutional Neural Network |
Sub Title (in English) | |
Keyword(1) | Multi-view 3D Image |
Keyword(2) | Super-resolution |
Keyword(3) | Chainer |
Keyword(4) | Convolutional Neural Network (CNN) |
Keyword(5) | H.265/HEVC |
Keyword(6) | Peak Signal to Noise Ratio (PSNR) |
1st Author's Name | Norifumi Kawabata |
1st Author's Affiliation | Nagoya University(Nagoya Univ.) |
Date | 2018-05-25 |
Paper # | IMQ2018-3 |
Volume (vol) | vol.118 |
Number (no) | IMQ-65 |
Page | pp.pp.15-20(IMQ), |
#Pages | 6 |
Date of Issue | 2018-05-18 (IMQ) |