Single image dehazing for a variety of haze scenarios using back projected pyramid network
This addresses the problem of robust image dehazing for computer vision applications, offering a method that works with minimal data, though it is incremental in improving existing techniques.
The paper tackles single image dehazing across diverse haze conditions, proposing BPPNet, a GAN architecture that achieves state-of-the-art performance on multiple NTIRE datasets using as few as 20 training image pairs.
Learning to dehaze single hazy images, especially using a small training dataset is quite challenging. We propose a novel generative adversarial network architecture for this problem, namely back projected pyramid network (BPPNet), that gives good performance for a variety of challenging haze conditions, including dense haze and inhomogeneous haze. Our architecture incorporates learning of multiple levels of complexities while retaining spatial context through iterative blocks of UNets and structural information of multiple scales through a novel pyramidal convolution block. These blocks together for the generator and are amenable to learning through back projection. We have shown that our network can be trained without over-fitting using as few as 20 image pairs of hazy and non-hazy images. We report the state of the art performances on NTIRE 2018 homogeneous haze datasets for indoor and outdoor images, NTIRE 2019 denseHaze dataset, and NTIRE 2020 non-homogeneous haze dataset.