Single image dehazing for a variety of haze scenarios using back projected pyramid network

TL;DR

This paper introduces BPPNet, a novel GAN architecture that effectively dehazes images across various haze conditions using minimal training data, achieving state-of-the-art results.

Contribution

The paper presents BPPNet, a new generative adversarial network with pyramidal convolution blocks and back projection, capable of learning from limited data for diverse haze scenarios.

Findings

Achieves state-of-the-art performance on multiple NTIRE haze datasets.

Effective with as few as 20 training image pairs.

Handles dense and inhomogeneous haze conditions robustly.

Abstract

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.