End-to-end Dehazing Neural Network

B. Li, X. Peng, Z. Wang, J. Xu, D. Feng

Woman in Computer Vision Workshop(WiCV), IEEE Conference on Computer Vision and Pattern Recognition (CVPR2017)

Full Paper Version

Abstract: We propose an image dehazing model built with a convolutional neural network (CNN), which can dehaze with end-to-end design. It is designed based on a re-formulated dehazing model. Instead of estimating the transmission matrix and the atmospheric light separately as most previous models did, our network directly generates the clean image through a light-weight CNN. Experimental results on both synthesized and natural hazy image datasets demonstrate our superior performance than the state-of-the-art in terms of PSNR, SSIM and the subjective visual quality.