Deep Convolution Networks for Compression Artifacts Reduction

TL;DR

This paper introduces a deep convolutional network designed to effectively reduce various compression artifacts in images, achieving high performance and speed suitable for real-world applications.

Contribution

The paper proposes a compact, efficient CNN framework for artifact reduction, incorporating layer decomposition and transfer learning, with significant speed improvements and superior results.

Findings

Achieves 7.5x speedup with minimal performance loss

Outperforms state-of-the-art methods on benchmarks

Effectively trains deeper models using transfer learning

Abstract

Lossy compression introduces complex compression artifacts, particularly blocking artifacts, ringing effects and blurring. Existing algorithms either focus on removing blocking artifacts and produce blurred output, or restore sharpened images that are accompanied with ringing effects. Inspired by the success of deep convolutional networks (DCN) on superresolution, we formulate a compact and efficient network for seamless attenuation of different compression artifacts. To meet the speed requirement of real-world applications, we further accelerate the proposed baseline model by layer decomposition and joint use of large-stride convolutional and deconvolutional layers. This also leads to a more general CNN framework that has a close relationship with the conventional Multi-Layer Perceptron (MLP). Finally, the modified network achieves a speed up of 7.5 times with almost no performance loss compared to the baseline model. We also demonstrate that a deeper model can be effectively trained with features learned in a shallow network. Following a similar "easy to hard" idea, we systematically investigate three practical transfer settings and show the effectiveness of transfer learning in low-level vision problems. Our method shows superior performance than the state-of-the-art methods both on benchmark datasets and a real-world use case.