Abstract
Most recent learning algorithms for single image dehazing are designed to train with paired hazy and corresponding ground truth images, typically synthesized images. Real paired datasets can help to improve performance, but are tough to acquire. This paper proposes an unsupervised dehazing algorithm based on GAN to alleviate this issue. An end-to-end network based on GAN architecture is established and fed with unpaired clean and hazy images, signifying that the estimation of atmospheric light and transmission is not required. The proposed network consists of three parts: a generator, a global test discriminator, and a local context discriminator. Moreover, a dark channel prior based attention mechanism is applied to handle inconsistency haze. We conduct experiments on RESIDE datasets. Extensive experiments demonstrated the effectiveness of the proposed approach which outperformed previous state-of-the-art unsupervised methods by a large margin.
Highlights
Haze is a traditional atmospheric phenomenon caused by small particles absorbing and scattering the light in the atmosphere
Adversarial loss aside, we introduce perceptual loss to assess the differences between Visual Geometry Group Network (VGG) features of hazy and dehazing images
We propose an attention mechanism inspired by dark channel prior to further process the sharply changing local area in the image
Summary
Haze is a traditional atmospheric phenomenon caused by small particles absorbing and scattering the light in the atmosphere. Images of outdoor scenes captured from hazy fields typically suffer from low contrast and poor visibility. An atmospheric scattering model was derived by McCarney [1] in 1975 to describe the haze mechanism. Several traditional methods [2,3,4,5,6,7,8] re proposed basing on this model. Single-image dehazing based on the atmospheric scattering model is an underconstrained problem that depends on an unknown depth and a highly ill-posed inverse problem. Traditional dehazing methods generally make additional assumptions and priors to restrict the model boundary and conditions. These assumptions may lose effectiveness and damage the quality of recovered image
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
