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

Short summary

A novel convolutional neural network (CNN) architecture extracts features in low-resolution (LR) space and uses an efficient sub-pixel convolution layer to upscale, achieving real-time 1080p video super-resolution on a K2 GPU.

AI-generated from the title and abstract; the full text is not read.