Shallow-UWnet : Compressed Model for Underwater Image Enhancement

TL;DR

This paper introduces Shallow-UWnet, a lightweight neural network for underwater image enhancement that maintains high performance while reducing computational and memory requirements, suitable for portable devices.

Contribution

The paper proposes a shallow neural network architecture that outperforms or matches deep models in underwater image enhancement with fewer parameters and better generalization.

Findings

Maintains performance with fewer parameters

Effective on both synthetic and real-world datasets

Suitable for deployment on portable devices

Abstract

Over the past few decades, underwater image enhancement has attracted increasing amount of research effort due to its significance in underwater robotics and ocean engineering. Research has evolved from implementing physics-based solutions to using very deep CNNs and GANs. However, these state-of-art algorithms are computationally expensive and memory intensive. This hinders their deployment on portable devices for underwater exploration tasks. These models are trained on either synthetic or limited real world datasets making them less practical in real-world scenarios. In this paper we propose a shallow neural network architecture, \textbf{Shallow-UWnet} which maintains performance and has fewer parameters than the state-of-art models. We also demonstrated the generalization of our model by benchmarking its performance on combination of synthetic and real-world datasets.