Performs row by row 1D convolutions of the given 2D image with the given 1D kernel.

Normalize all elements of the given numpy array to [0,1]

Smooth given col by averaging + pixel shuffle

Concat 4 flipped copies of input gradients (makes them wrap). Output is twice bigger in both dimensions.

Generates the corner part of the pyramidal-like mask. Currently, only for square shapes.

Downscale image by a factor of 2

Frankot-Chellappa depth-from-gradient algorithm.

Returns a 1D gaussian kernel of size 'length'.

Generates a pyramidal-like mask. Used for mixing overlapping predicted tiles.

Applies smoothing to tiles edges

Merges the list of tiles given an upscale factor and without overlap. img_size is the original size, before upscale & padding.

Returns list of tiles from the given image and the padding used to fit the tiles in it. Input image must have dimension C,H,W.

Upscale image. 'color_img' must be a numpy array in C,H,W format (with C as RGB). 'factor'' must be 'x2' or 'x4'.

Computes a height map from the given normal map. 'normals_img' must be a numpy array in C,H,W format (with C as RGB). 'seamless' is a bool that s

Computes a curvature map from the given normal map. 'normals_img' must be a numpy array in C,H,W format (with C as RGB). 'blur_radius' must be on

Computes a normal map from the given color map. 'color_img' must be a numpy array in C,H,W format (with C as RGB). 'overlap' must be one of 'SMAL

Converts a Blender image to a numpy C,H,W numpy array.

Normalize each pixel to unit vector.

Converts a C,H,W numpy image array to an array of pixel suited for blender internal images pixels.

Infer each tile with the given model. progress_callback will be called with arguments : current tile idx and total tiles amount (used to show pro

Merges the list of tiles into one image. img_size is the original size, before padding.

Returns list of tiles from the given image and the padding used to fit the tiles in it. Input image must have dimension C,H,W.