Apparatus and method of efficient sample adaptive offset
A method and apparatus for sample adaptive offset (SAO) compensation of reconstructed video data are disclosed. In one embodiment, the relation between the current pixel and said one or more neighboring pixels is stored so that the SAO compensated current pixel can replace the current pixel without buffering the to-be-processed pixels for classification. The SAO process may be performed on a region by region basis to adapt to the local characteristics of the picture.
1. A method of sample adaptive offset (SAO) compensation of reconstructed video data, the method comprising:
receiving first video data corresponding to the reconstructed video data or reconstructed-and-deblocked video data;
determining a category for a current pixel of the first video data according to a pixel intensity relation between the current pixel and one or more neighboring pixels, wherein the relation between the current pixel and said one or more neighboring pixels is based on a sign function;
storing the pixel intensity relation between the current pixel and said one or more neighboring pixels;
compensating the current pixel using an offset value associated with the category to generate a compensated current pixel; and
replacing the current pixel of the first video data by the compensated current pixel after said storing the pixel intensity relation between the current pixel and said one or more neighboring pixels without a need for completion of compensating all of said one or more neighboring pixels.
2. The method of claim 1 , wherein at least a portion of the relation between the current pixel and said one or more neighboring pixels is used during determining a category of another pixel belongs to said one or more neighboring pixels.
3. The method of claim 1 , wherein a look-up table is used for determining the category for the current pixel.
4. An apparatus for sample adaptive offset (SAO) compensation of reconstructed video data, the apparatus comprising one or more electronic circuits configured to:
receive first video data corresponding to the reconstructed video data or reconstructed-and-deblocked video data;
determine a category for a current pixel of the first video data according to pixel intensity relation between the current pixel and one or more neighboring pixels, wherein the relation between the current pixel and said one or more neighboring pixels is based on a sign function;
store the pixel intensity relation between the current pixel and said one or more neighboring pixels;
compensate the current pixel using an offset value associated with the category to generate a compensated current pixel; and
replace the current pixel of the first video data by the compensated current pixel after the pixel intensity relation between the current pixel and said one or more neighboring pixels is stored without a need for completion of compensating all of said one or more neighboring pixels.
5. The apparatus of claim 4 , wherein at least a portion of the relation between the current pixel and said one or more neighboring pixels is used during determining a category of another pixel belongs to said one or more neighboring pixels.
6. The apparatus of claim 4 , wherein a look-up table is used for determining the category for the current pixel.
7. A method of sample adaptive offset (SAO) compensation of video data, the method comprising:
receiving first video data corresponding to one of: reconstructed video data and reconstructed-and-deblocked video data;
determining and storing a pixel intensity relation between a current pixel of the first video data and at least one neighboring pixel, wherein the pixel intensity relation is determined based on a sign of a difference between the current pixel and each of the at least one neighboring pixel;
determining a category for the current pixel according to the pixel intensity relation;
applying an offset value to the current pixel to generate a compensated current pixel, the offset value being associated with the determined category; and
replacing the current pixel of the first video data with the compensated current pixel upon storing the pixel intensity relation without a need for completion of compensating all of said one or more neighboring pixels.