ImageModel

Computes the image model from an in_gray image.

The outputs are a low-pass filtered version of in_gray using a 3x3 Gaussian filter, and a 2D image representing the XY partial differences of the low-pass image.

Parameters

float_precision : int. Defaults to ll.FloatPrecision.FP32. Floating point precision used accross the algorithm. The outputs out_gray and out_gradient will be of this floating point precision.

Inputs

in_gray : ImageView. r8ui image.

Outputs

out_gray : ImageView {r16f, r32f} image. The low-pass filtered version of in_gray. The values are normalized to the range [0, 1]

out_gradient: ImageView {rg16f, rg32f} image. The X and Y gradient components of in_gray. The values are normalized to the range [-1, 1]