DScCP

class deepinv.models.DScCP(K=20, F=64, pretrained='download', device=None)

Bases: Module

DScCP denoiser network.

The network architecture is based on the paper Unfolded proximal neural networks for robust image Gaussian denoising, and has an unrolled architecture based on the fast Chambolle-Pock algorithm using strong convexity. DScCP stands for Deep Strongly Convex Chambolle Pock.

The pretrained weights are trained with the default parameters of the network, i.e. K=20 layers, F=64 channels. They can be downloaded via setting pretrained='download'.

Parameters:

• K (int) – depth i.e. number of convolutional layers.

• F (int) – number of channels per convolutional layer.

• pretrained (str) – ‘download’ to download pretrained weights, or path to local weights file.

• device (torch.device, str) – ‘cuda’ or ‘cpu’.

forward(x, sigma=0.03)

Run the denoiser on noisy image.

Parameters:

• x (torch.Tensor) – noisy image.

• sigma (float) – noise level.