redR: REgularization by Denoising (RED)
Regularization by Denoising uses a denoising engine to solve
many image reconstruction ill-posed inverse problems. This is a R
implementation of the algorithm developed by Romano et.al. (2016) <arXiv:1611.02862>. Currently,
only the gradient descent optimization framework is implemented. Also,
only the median filter is implemented as a denoiser engine. However,
(almost) any denoiser engine can be plugged in. There are currently available
3 reconstruction tasks: denoise, deblur and super-resolution. And again,
any other task can be easily plugged into the main function 'RED'.
||R (≥ 3.4.0), imager
||Adriano Passos [aut, cre]
||Adriano Passos <adriano.utfpr at gmail.com>
Please use the canonical form
to link to this page.