EE Seminar: NLDNet++: A Physics Based Single Image Dehazing Network

27 בינואר 2020, 15:00 
Room 011, Kitot Building 

Speaker: Iris Tal

M.Sc. student under the supervision of Prof. Shai Avidan

 

Monday, January 27, 2020 at 15:00

Room 011, Kitot Bldg., Faculty of Engineering

 

NLDNet++: A Physics Based Single Image Dehazing Network

Abstract

Deep learning methods for image dehazing achieve impressive results. Yet, the task of collecting ground truth hazy/dehazed image pairs to train the network is cumbersome. We propose to use Non-Local Image Dehazing (NLD), an existing physics based technique, to provide the dehazed image required to training a network.

Upon close inspection, we find that NLD suffers from several shortcomings and propose novel extensions to improve it. The new method, termed NLD++, consists of 1) denoising the input image as pre-processing step to avoid noise amplification, 2) introducing a constrained optimization that respects physical constraints. NLD++ produces superior results to NLD at the expense of increased computational cost. To offset that, we propose NLDNet++, a fully convolutional network that is trained on pairs of hazy images and images dehazed by NLD++. This eliminates the need of existing deep learning methods that require hazy/dehazed image pairs that are difficult to obtain. We evaluate the performance of NLDNet++ on standard data sets and find it to compare favorably with existing methods.

אוניברסיטת תל-אביב, ת.ד. 39040, תל-אביב 6997801
UI/UX Basch_Interactive