Feature Fusion Attention Network with CycleGAN for Image Dehazing, De-Snowing and De-Raining
This addresses image quality restoration for computer vision applications, but it is incremental as it builds on existing techniques.
The paper tackled image dehazing by combining Feature Fusion Attention networks with CycleGAN, achieving superior PSNR and SSIM scores compared to traditional methods.
This paper presents a novel approach to image dehazing by combining Feature Fusion Attention (FFA) networks with CycleGAN architecture. Our method leverages both supervised and unsupervised learning techniques to effectively remove haze from images while preserving crucial image details. The proposed hybrid architecture demonstrates significant improvements in image quality metrics, achieving superior PSNR and SSIM scores compared to traditional dehazing methods. Through extensive experimentation on the RESIDE and DenseHaze CVPR 2019 dataset, we show that our approach effectively handles both synthetic and real-world hazy images. CycleGAN handles the unpaired nature of hazy and clean images effectively, enabling the model to learn mappings even without paired data.