GAMA-IR: Global Additive Multidimensional Averaging for Fast Image Restoration

TL;DR

This paper introduces a fast, shallow image restoration network using global additive multidimensional averaging, achieving state-of-the-art results with significantly reduced latency and computational cost.

Contribution

The proposed GAMA-IR network is a novel shallow architecture that captures global information efficiently, outperforming deeper models in speed while maintaining high image restoration quality.

Findings

Exceeds SIDD denoising state-of-the-art by 0.11dB

Achieves 2 to 10 times faster processing than existing methods

Maintains comparable or superior performance with less latency

Abstract

Deep learning-based methods have shown remarkable success for various image restoration tasks such as denoising and deblurring. The current state-of-the-art networks are relatively deep and utilize (variants of) self attention mechanisms. Those networks are significantly slower than shallow convolutional networks, which however perform worse. In this paper, we introduce an image restoration network that is both fast and yields excellent image quality. The network is designed to minimize the latency and memory consumption when executed on a standard GPU, while maintaining state-of-the-art performance. The network is a simple shallow network with an efficient block that implements global additive multidimensional averaging operations. This block can capture global information and enable a large receptive field even when used in shallow networks with minimal computational overhead. Through extensive experiments and evaluations on diverse tasks, we demonstrate that our network achieves comparable or even superior results to existing state-of-the-art image restoration networks with less latency. For instance, we exceed the state-of-the-art result on real-world SIDD denoising by 0.11dB, while being 2 to 10 times faster.