Single Image Haze Removal using Dark Channel Prior and Guided Image Filtering/Fast Guided Filter

Numerical Imaging Project - MVA 2020/2021

Reimplementation of the Haze Removal method Single Image Haze Removal Using Dark Channel Prior from He et al. using Guided Image Filtering from He et al. and optimizing the computation with Fast Guided Filter (He et al.).

To dehaze a single image, simply run python main.py -p path/to/image --resize max_size_you_want from the root folder.
To compare all the matting methods on a single image, run python evaluate_matting.py -p path/to/image -s path/to/save_folder --resize max_size_you_want from the benchmarks folder.
To compute all the parameters evaluations on a whole folder, run python launch_all_evaluations.py -f path/to/images/folder -s path/to/save/folder --resize max_size_you_want from the benchmarks folder.

K. He, J. Sun, and X. Tang. Single Image Haze Removal Using Dark Channel Prior. IEEETransactions on Pattern Analysis and Machine Intelligence, 33(12):2341–2353, 2011
K. He, J. Sun, and X. Tang. Guided Image Filtering. IEEE Transactions on PatternAnalysis and Machine Intelligence, 35(6):1397–1409, 2013
Kaiming He and Jian Sun. Fast Guided Filter. ArXiv, abs/1505.00996, 2015