PT Unknown
AU Patricia Suarez
   Angel Sappa
   Boris X. Vintimilla
   Riad I. Hammoud
TI Deep Learning based Single Image Dehazing
BT 31st IEEE Conference on Computer Vision and Pattern Recognition Workhsop
PY 2018
BP 1250
EP 12507
DI 10.1109/CVPRW.2018.00162
DE Gallium nitride; Atmospheric modeling; Generators; Generative adversarial networks; Convergence; Image color analysis
AB This paper proposes a novel approach to remove haze degradations in RGB images using a stacked conditional Generative Adversarial Network (GAN). It employs a triplet of GAN to remove the haze on each color channel independently.A multiple loss functions scheme, applied over a conditional probabilistic model, is proposed. The proposed GAN architecture learns to remove the haze, using as conditioned entrance, the images with haze from which the clearimages will be obtained. Such formulation ensures a fast model training convergence and a homogeneous model generalization. Experiments showed that the proposed method generates high-quality clear images.
ER