A Light Dual-Task Neural Network for Haze Removal
This addresses the problem of image quality degradation due to haze for computer vision applications, representing an incremental improvement by reducing reliance on artificial priors.
The paper tackles single-image dehazing by proposing LDTNet, a light dual-task neural network that restores haze-free images in one shot, achieving superior performance against state-of-the-art methods on synthetic and real-world images.
Single-image dehazing is a challenging problem due to its ill-posed nature. Existing methods rely on a suboptimal two-step approach, where an intermediate product like a depth map is estimated, based on which the haze-free image is subsequently generated using an artificial prior formula. In this paper, we propose a light dual-task Neural Network called LDTNet that restores the haze-free image in one shot. We use transmission map estimation as an auxiliary task to assist the main task, haze removal, in feature extraction and to enhance the generalization of the network. In LDTNet, the haze-free image and the transmission map are produced simultaneously. As a result, the artificial prior is reduced to the smallest extent. Extensive experiments demonstrate that our algorithm achieves superior performance against the state-of-the-art methods on both synthetic and real-world images.