Generic 3D Convolutional Fusion for image restoration

TL;DR

This paper introduces a versatile 3D convolutional fusion method that improves image denoising and super-resolution performance using a unified architecture, achieving significant PSNR gains while maintaining efficiency.

Contribution

The novel 3D convolutional fusion (3DCF) approach unifies image denoising and super-resolution tasks within a single network architecture, enhancing performance over state-of-the-art methods.

Findings

Achieves 0.1dB-0.4dB PSNR improvement over existing methods.

Effective for both denoising and super-resolution tasks.

Maintains computational efficiency.

Abstract

Also recently, exciting strides forward have been made in the area of image restoration, particularly for image denoising and single image super-resolution. Deep learning techniques contributed to this significantly. The top methods differ in their formulations and assumptions, so even if their average performance may be similar, some work better on certain image types and image regions than others. This complementarity motivated us to propose a novel 3D convolutional fusion (3DCF) method. Unlike other methods adapted to different tasks, our method uses the exact same convolutional network architecture to address both image denois- ing and single image super-resolution. As a result, our 3DCF method achieves substantial improvements (0.1dB-0.4dB PSNR) over the state-of-the-art methods that it fuses, and this on standard benchmarks for both tasks. At the same time, the method still is computationally efficient.