up
100
作者 blueardour 2019-04-30 02:38:59
Wrote 0 BlogsTotally 0 words
Super resolution

Introduction

Super resolution (SR) is some task in which the input is a low resolution image or blur image and a high resolution or deblur image is required to output. Some researchers solve the problem using deep learning based method. For the super resolution scenario, it could be divided in to single image SR (SISR), multiple frame images SR (MISR) and video SR.

SISR

Survey

Deep Learning for Single Image Super-Resolution: A Brief Review

SRCNN / FSRCNN

Accurate Image Super-Resolution Using Very Deep Convolutional Networks (VDSR)

Paper link
on CVPR 2016 oral

Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network (ESPCN)

Paper link
on CVPR 2016

Enhanced Deep Residual Networks for Single Image Super-Resolution (EDSR/MDSR)

Paper link
Code
on CVPR 2017

Wide Activation for Efficient and Accurate Image Super-Resolution

Paper link

Single Image Super-Resolution via Squeeze and Excitation Network

Paper link

Video SR

Detail-revealing Deep Video Super-resolution

Paper link

-->