Fast Image Processing with Fully-Convolutional Networks

TL;DR

This paper introduces a fully-convolutional neural network approach to approximate and accelerate various image processing operators, achieving high accuracy and real-time performance without executing the original operators.

Contribution

The paper presents a novel fully-convolutional network architecture that efficiently approximates multiple image processing operators, outperforming prior methods in accuracy and speed.

Findings

8.5 dB PSNR improvement over previous methods

3x reduction in DSSIM compared to prior schemes

Operates in constant time at full resolution

Abstract

We present an approach to accelerating a wide variety of image processing operators. Our approach uses a fully-convolutional network that is trained on input-output pairs that demonstrate the operator's action. After training, the original operator need not be run at all. The trained network operates at full resolution and runs in constant time. We investigate the effect of network architecture on approximation accuracy, runtime, and memory footprint, and identify a specific architecture that balances these considerations. We evaluate the presented approach on ten advanced image processing operators, including multiple variational models, multiscale tone and detail manipulation, photographic style transfer, nonlocal dehazing, and nonphotorealistic stylization. All operators are approximated by the same model. Experiments demonstrate that the presented approach is significantly more accurate than prior approximation schemes. It increases approximation accuracy as measured by PSNR across the evaluated operators by 8.5 dB on the MIT-Adobe dataset (from 27.5 to 36 dB) and reduces DSSIM by a multiplicative factor of 3 compared to the most accurate prior approximation scheme, while being the fastest. We show that our models generalize across datasets and across resolutions, and investigate a number of extensions of the presented approach. The results are shown in the supplementary video at https://youtu.be/eQyfHgLx8Dc