ImageProcessor
Computes the parameters from the input image sequence.
Computes the parameters from the input image sequence.
The layout of the out_image_params
vector is as follows:
out_image_params.xy
: The X and Y gradient components of in_gray. The values are normalized to the range [-1, 1]out_image_params.z
: The time difference (in_gray - in_gray_old)out_image_params.w
: The regularization value1.0 / (alpha_square + grad_x*grad_x + grad_y*grad_y)
Parameters
alpha : float. Defaults to 0.05. Regularization gain.
float_precision : int. Defaults to ll.FloatPrecision.FP32. Floating point precision used accross the algorithm. The out_image_params output will be of this floating point precision.
Inputs
in_gray : ImageView. r8ui image.
in_gray_old: ImageView. {r16f, r32f} image. The low-pass filtered version of in_gray. The values are normalized to the range [0, 1].
Outputs
out_image_params : ImageView {rgba16f, rgba32f} image. Output image parameters
Feedback
Was this page helpful?
Glad to hear it! Please tell us how we can improve.
Sorry to hear that. Please tell us how we can improve.