A New Real-World Video Dataset for the Comparison of Defogging Algorithms

TL;DR

This paper introduces VIREDA, a new real-world video dataset with varying fog densities and clear ground truths, to benchmark and develop video defogging algorithms, emphasizing the potential of Transformer-based models.

Contribution

The paper presents VIREDA, a novel dataset for real-world video defogging, and explores Transformer architectures for improved defogging performance.

Findings

VIREDA provides diverse foggy videos with ground truths for benchmarking.

Transformer-based models show promising results on the VIREDA dataset.

The dataset facilitates development of more effective video defogging algorithms.

Abstract

Video restoration for noise removal, deblurring or super-resolution is attracting more and more attention in the fields of image processing and computer vision. Works on video restoration with data-driven approaches for fog removal are rare however, due to the lack of datasets containing videos in both clear and foggy conditions which are required for deep learning and benchmarking. A new dataset, called REVIDE, was recently proposed for just that purpose. In this paper, we implement the same approach by proposing a new REal-world VIdeo dataset for the comparison of Defogging Algorithms (VIREDA), with various fog densities and ground truths without fog. This small database can serve as a test base for defogging algorithms. A video defogging algorithm is also mentioned (still under development), with the key idea of using temporal redundancy to minimize artefacts and exposure variations between frames. Inspired by the success of Transformers architecture in deep learning for various applications, we select this kind of architecture in a neural network to show the relevance of the proposed dataset.