| 摘要 |
In one embodiment, a method for processing a video signal includes: (1) receiving and storing luminance and chrominance information for each pixel in a first portion of the signal; (2) receiving luminance and chrominance information for each pixel in a second portion of the signal; (3) determining an estimated motion vector for each particular pixel of the second portion by comparing the luminance and chrominance information of the particular pixel to the stored luminance and chrominance information for one or more pixels in a search area of the first portion to determine a pixel in the search area that most closely matches the particular pixel and determining the estimated motion vector according to the particular pixel and the most closely matching pixel; (4) using the estimated motion vector to access the chrominance information for the most closely matching pixel; (5) using a three-dimensional comb filter to filter the chrominance information for the particular pixel and for the most closely matching pixel to determine three-dimensional filter output chrominance information for the particular pixel; (6) receiving the three-dimensional output chrominance information and if the chrominance information for the particular pixel matches the chrominance information of the most closely matching pixel within a specified tolerance, using the three-dimensional filter output chrominance information to determine output luminance information for the particular pixel; and (7) providing the three-dimensional output chrominance information and output luminance information for the particular pixel for displaying the particular pixel.
|
