TH44167B - Low-complexity film grain simulation technique - Google Patents

Abstract

DC60 The simulation of the film grain in the video image occurs by forming the first group (ie, the matrix sequence) of the transformed coefficients. For a series of fHL, fVL, fHH and fVH cut-off frequencies in combination with the desired film grain pattern (fHL, fVL, fHH and fVH cut-off frequencies represent the two-dimensional cut-off frequency of the filter characterizing the film grain effect. The coefficients being converted are inverted to obtain a bit-sized accurate sample of the film grain. And example that Precisely, the bit size is scaled so that it can be mixed with a video signal to simulate film grain in the signal. The simulation of film grain in a video image occurs by first forming a cluster (that is, a matrix array) of Converted coefficients For a series of fHL, fVL, fHH and fVH cut-off frequencies in combination with the desired film grain pattern (fHL, fVL, fHH and fVH cut-off frequencies represent the two-dimensional cut-off frequency of the filter characterizing the film grain effect. The coefficients being converted are inverted to obtain a bit-sized accurate sample of the film grain. And example that Precisely, the bit sizes are scaled so that they can be mixed with a video signal to simulate film grain in the signal.

Claims (1)

1. The method for simulating film grain patterns consists of the following steps: constructing a transformation coefficient for the film grain sample, where the coefficient is random as long as it lies within the band-pass region defined by a set of cutoff frequencies characteristic of the desired film grain pattern (Step 202), and constructing a reverse transformation for the constructed coefficient to obtain the resulting film grain sample (Step 204).