TVPrior
- class deepinv.optim.TVPrior(def_crit=1e-08, n_it_max=1000, *args, **kwargs)[source]
Bases:
Prior
Total variation (TV) prior \(\reg{x} = \| D x \|_{1,2}\).
- Parameters:
- fn(x, *args, **kwargs)[source]
Computes the regularizer
\[\reg{x} = \|Dx\|_{1,2}\]where D is the finite differences linear operator, and the 2-norm is taken on the dimension of the differences.
- Parameters:
x (torch.Tensor) – Variable \(x\) at which the prior is computed.
- Returns:
(torch.Tensor) prior \(g(x)\).
- nabla(x)[source]
Applies the finite differences operator associated with tensors of the same shape as x.
- prox(x, *args, gamma=1.0, **kwargs)[source]
Compute the proximity operator of TV with the denoiser
TVDenoiser
.- Parameters:
x (torch.Tensor) – Variable \(x\) at which the proximity operator is computed.
gamma (float) – stepsize of the proximity operator.
- Returns:
(torch.Tensor) proximity operator at \(x\).
Examples using TVPrior
:
Image deblurring with Total-Variation (TV) prior
Image deblurring with Total-Variation (TV) prior