Deep Networks for Image and Video Super-Resolution

TL;DR

This paper introduces a novel deep convolutional network architecture with mixed-dense blocks for single image and video super-resolution, achieving improved performance and efficiency over existing methods.

Contribution

The paper proposes a new deep network with mixed-dense connection blocks and a scale-recurrent framework for efficient multi-scale super-resolution.

Findings

Outperforms state-of-the-art on image super-resolution benchmarks.

Achieves better spatio-temporal consistency in video super-resolution.

Demonstrates improved quality with different loss configurations.

Abstract

Efficiency of gradient propagation in intermediate layers of convolutional neural networks is of key importance for super-resolution task. To this end, we propose a deep architecture for single image super-resolution (SISR), which is built using efficient convolutional units we refer to as mixed-dense connection blocks (MDCB). The design of MDCB combines the strengths of both residual and dense connection strategies, while overcoming their limitations. To enable super-resolution for multiple factors, we propose a scale-recurrent framework which reutilizes the filters learnt for lower scale factors recursively for higher factors. This leads to improved performance and promotes parametric efficiency for higher factors. We train two versions of our network to enhance complementary image qualities using different loss configurations. We further employ our network for video super-resolution task, where our network learns to aggregate information from multiple frames and maintain spatio-temporal consistency. The proposed networks lead to qualitative and quantitative improvements over state-of-the-art techniques on image and video super-resolution benchmarks.