Sorry, you need to enable JavaScript to visit this website.

DENSE BYNET: RESIDUAL DENSE NETWORK FOR IMAGE SUPER RESOLUTION

Citation Author(s):
Jiu Xu, Yeongnam Chae, Bjorn Stenger, Ankur Datta
Submitted by:
Yeongnam Chae
Last updated:
5 October 2018 - 1:41am
Document Type:
Presentation Slides
Document Year:
2018
Event:
Presenters:
Yeongnam Chae
Paper Code:
2118

Abstract

This paper proposes a method, Dense ByNet, for single image super-resolution based on a convolutional neural network (CNN). The main innovation is a new architecture that combines several CNN design choices. Using a residual network as a basis, it introduces dense connections inside residual blocks, significantly reducing the number of parameters. Second, we apply dilation convolutions to increase the spatial context. Lastly, we propose modifications to the activation and cost functions. We evaluate the method on benchmark datasets and show that it achieves state-of-the-art results over multiple upscaling factors in terms of peak SNR and structural similarity (SSIM).

up
0 users have voted:

Files

NEW_R_Dense ByNet.pdf

(225)