Make coordinates at grid centers.

Query RGB values of each coordinate using latent modulations and latent codes. (without CMSR) :param coef: Coefficient (B, C, h, w)

Query RGB values of each coordinate using latent modulations and latent codes. (without CMSR) :param feat: Feature maps (B, C, h, w)

Query RGB values of each coordinate using latent modulations and latent codes. (without CMSR) :param feat: Feature maps (B, C, h, w)

Convert the image to coord-RGB pairs. img: Tensor, (3, H, W)

Perform batched predictions using a model on provided inputs. :param model: The model used for generating predictions. :param inp: Input

Query RGB values of each coordinate batch using latent modulations and latent codes. :param coef: Coefficient (B, C, h, w) :

Query RGB values of each coordinate batch using latent modulations and latent codes. :param feat: Feature maps (B, C, h, w)

Batched predict. Args: feat (Tensor): Input tensor. coord (Tensor): coord tensor. cell (Tensor): cell ten

Calculate the box coordinates to crop an image at its center. :param w: Width of the original image. :param h: Height of the original im

Evaluate the Peak Signal-to-Noise Ratio (PSNR) of a model over a dataset loaded through a specified loader. :param loader: The DataLoader pr

Extract patches from images and put them in the C output dimension. :param padding: :param images: [batch, channels, in_rows, in_cols]. A

Patch matching between input and reference features. Args: feat_input (Tensor): the feature of input, shape: (c, h, w). feat_ref (

Query RGB value of GT. Copyright (c) 2020, Yinbo Chen, under BSD 3-Clause License. Args: feature (Tensor): encoded featu

Extract sliding local patches from an input feature tensor. The sampled pathes are row-major. Args: inputs (Tensor): the input feature

Tensor shift. Args: x (Tensor): the input tensor. The shape is [b, h, w, c]. shift (tuple): shift pixel. fill_val (float)

Args: x: (B, H, W, C) window_size (int): window size Returns: windows: (num_windows*B, window_size, window_size, C)

Batched predict. Args: x (Tensor): Input tensor. coord (Tensor): coord tensor. cell (Tensor): cell tensor

:param loader: :param model: :param scale: :param data_norm: :param window_size: :param trained_scale: :return:

Generate latent codes using the encoder. :param inp: Input image (B, h * w, 3) :param inp_coord: Input coordinates (B, h * w

Generate latent codes using the encoder. :param inp: Input image (B, h * w, 3) :param inp_coord: Input coordinates (B, h * w

Generate latent codes using the encoder. :param inp: Input image (B, h * w, 3) :param inp_coord: Input coordinates (B, h * w

Generate latent modulations using the latent MLP. :param coef: Coefficient (B, C, h, w) :param freq: Frequency (B, C, h, w)

Generate latent modulations using the latent MLP. :param feat: Feature maps (B, C, h, w) :param cell: Cell areas (B, H * W,

Inserts bn layer after each conv. Args: names (list): The list of layer names. Returns: list: The list of layer names with b

Prepare the coordinates and cells. :param feat: Latent modulations (B, C', h, w) :param coord: Coordinates (B, H * W, 2)

:param feat: Feature maps (B, C, h, w) :param scale: Cell areas (B, H * W, 2) :return:

Query RGB values of each coordinate using latent modulations and latent codes. (CMSR included) :param coef: Coefficient (B, C, h, w)

Query RGB values of each coordinate using latent modulations and latent codes. (CMSR included) :param feat: Feature maps (B, C, h, w

Query RGB values of each coordinate using latent modulations and latent codes. (CMSR included) :param feat: Feature maps (B, C, h, w

Query RGB values of input coordinates using latent modulations and latent codes. :param coef: Coefficient (B, C, h, w) :para

Query RGB values of each coordinate using latent modulations and latent codes. (without CMSR) :param feat: Feature maps (B, C, h, w)

Query RGB values of each coordinate using latent modulations and latent codes. (without CMSR) :param feat: Feature maps (B, C, h, w)

Resize all images in a folder by a specified scale factor using a specified interpolation method. :param folder: Directory containing images

Update the Scale2mod table for CMSR testing. :param coef: Coefficient (B, C, h, w) :param freq: Frequency (B, C, h, w)

Update the Scale2mod table for CMSR testing. :param feat: Feature maps (B, C, h, w) :param mod: Latent modulations (B, C', h

Update the Scale2mod table for CMSR testing. :param feat: Feature maps (B, C, h, w) :param mod: Latent modulations (B, C', h

Args: windows: (num_windows*B, window_size, window_size, C) window_size (int): Window size H (int): Height of image