NTIRE Workshop and Challenges @ CVPR 2018
Outdoor Image Dehazing Challenge
Haze removal or dehazing is a challenging ill-posed problem that has drawn a significant attention in the last few years. Despite this growing interest, the scientific community is still lacking a reference dataset to evaluate objectively and quantitatively the performance of proposed dehazing methods. The few datasets that are currently considered, both for assessment and training of learning-based dehazing techniques, exclusively rely on synthetic hazy images. To address this limitation, we introduce the first outdoor scenes database (named O-HAZE) composed of pairs of real hazy and corresponding haze-free images. In practice, hazy images have been captured in presence of real haze, generated by professional haze machines, and O-HAZE contains 45 different outdoor scenes depicting the same visual content recorded in haze-free and hazy conditions, under the same illumination parameters. To illustrate its usefulness, O-HAZE is used to compare a representative set of state-of-the-art dehazing techniques, using traditional image quality metrics such as PSNR, SSIM and CIEDE2000. This reveals the limitations of current techniques, and questions some of their underlying assumptions.
NTIRE is a CVPR workshop that aims to provide an overview of the new trends and advances in those areas. Moreover, it offers an opportunity for academic and industrial attendees to interact and explore collaborations. Jointly with workshop NTIRE organised in 2018 the first image dehazing online challenge.
O-HAZE has been employed in the dehazing challenge of the NTIRE 2018 CVPR workshop.
O-HAZE: a dehazing benchmark with real hazy and haze-free outdoor images
Bibtex
@inproceedings{O-HAZE_2018,
author = { Codruta O. Ancuti and Cosmin Ancuti and Radu Timofte and Christophe De Vleeschouwer},
title = {O-HAZE: a dehazing benchmark with real hazy and haze-free outdoor images},
booktitle = {IEEE Conference on Computer Vision and Pattern Recognition, NTIRE Workshop },
series = {NTIRE CVPR'18},
year = {2018},
location = {Salt Lake City, Utah, USA},
}
To avoid the parameter tweaking of some of the tested methods, in the paper all the resuls have been generated using the small size of the images (we kept the original image proportion and the images have been resized to a maximum size of 800 pixels).
