# Dense Haze: A benchmark for image dehazing with dense-haze and haze-free images

## TL;DR

This paper introduces Dense-Haze, a new dataset with real hazy and haze-free image pairs to improve the validation and development of single-image dehazing methods, revealing current techniques' limitations.

## Contribution

The creation of Dense-Haze, a dataset with real dense haze and corresponding haze-free images, and a comprehensive evaluation of existing dehazing methods on this dataset.

## Key findings

- Existing dehazing methods perform poorly on dense homogeneous haze.
- Dense-Haze dataset reveals significant room for improvement in current techniques.
- Real haze generation ensures realistic evaluation scenarios.

## Abstract

Single image dehazing is an ill-posed problem that has recently drawn important attention. Despite the significant increase in interest shown for dehazing over the past few years, the validation of the dehazing methods remains largely unsatisfactory, due to the lack of pairs of real hazy and corresponding haze-free reference images. To address this limitation, we introduce Dense-Haze - a novel dehazing dataset. Characterized by dense and homogeneous hazy scenes, Dense-Haze contains 33 pairs of real hazy and corresponding haze-free images of various outdoor scenes. The hazy scenes have been recorded by introducing real haze, generated by professional haze machines. The hazy and haze-free corresponding scenes contain the same visual content captured under the same illumination parameters. Dense-Haze dataset aims to push significantly the state-of-the-art in single-image dehazing by promoting robust methods for real and various hazy scenes. We also provide a comprehensive qualitative and quantitative evaluation of state-of-the-art single image dehazing techniques based on the Dense-Haze dataset. Not surprisingly, our study reveals that the existing dehazing techniques perform poorly for dense homogeneous hazy scenes and that there is still much room for improvement.