Algorithm

Recursion (computer science)

Gradient descent

Computer science

Resolution (logic)

Image resolution

Mathematics

Image (mathematics)

Artificial intelligence

Geometry

Connection (principal bundle)

Artificial neural network

Margin (machine learning)

Convolutional neural network

Machine learning

Computer vision

Convolution (computer science)

We propose an image super-resolution method (SR) using a deeply-recursive convolutional network (DRCN). Our network has a very deep recursive layer (up to 16 recursions). Increasing recursion depth can improve performance without introducing new parameters for additional convolutions. Albeit advantages, learning a DRCN is very hard with a standard gradient descent method due to exploding/ vanishing gradients. To ease the difficulty of training, we propose two extensions: recursive-supervision and skip-connection. Our method outperforms previous methods by a large margin.

Deeply-Recursive Convolutional Network for Image Super-Resolution