Marine Snow Removal Benchmarking Dataset

TL;DR

This paper presents a new publicly available dataset for benchmarking marine snow removal in underwater images, including modeled artifacts, synthesized image pairs, and initial benchmarking results to advance underwater image restoration.

Contribution

It introduces a comprehensive dataset with modeled marine snow artifacts and provides the first benchmarking results for marine snow removal methods.

Findings

First benchmarking results for marine snow removal

Synthesized datasets enable objective evaluation

Modeling of marine snow artifacts from real images

Abstract

This paper introduces a new benchmarking dataset for marine snow removal of underwater images. Marine snow is one of the main degradation sources of underwater images that are caused by small particles, e.g., organic matter and sand, between the underwater scene and photosensors. We mathematically model two typical types of marine snow from the observations of real underwater images. The modeled artifacts are synthesized with underwater images to construct large-scale pairs of ground truth and degraded images to calculate objective qualities for marine snow removal and to train a deep neural network. We propose two marine snow removal tasks using the dataset and show the first benchmarking results of marine snow removal. The Marine Snow Removal Benchmarking Dataset is publicly available online.