Method of selecting quantizer values in video compression systems
View Patent ↗A method of obtaining a quantization value for macroblocks of a picture for video compression includes estimating 11 , for a plurality of quantization values, a size of the picture in terms of a sum of all quantized coded coefficients. A linear relationship between a size of a macroblock and Activity of the macroblocks is assumed 12 , where the Activity of the macroblock is the number of coded coefficients in the macroblock before discarding any coefficients or quantizing the remaining coefficients. The macroblocks are divided 13 into groups by their Activity level such that substantially a same number of pre-quantization coefficients are in each group. An average Activity of each group is determined 14 . A quantization value for each group is determined 15 from the assumed linear relationship between a number of bits per macroblock and Activity of the macroblocks and the average Activity of the group, as imposed by a predetermined bit rate.
1. A method of obtaining a quantisation value for macroblocks of a picture in a video compression system comprising the steps of:
a. estimating, in a processor, sizes in the transform domain of each macroblock of a picture in terms of a sum of all coded coefficients for a plurality of quantisation values;
b. assuming, in the processor, a relationship between a size of a macroblock and Activity of the macroblock for each quantization value;
c. dividing, in the processor, macroblocks into groups by Activity;
d. determining, in the processor, an average Activity of each group; and
e. calculating, in the processor, a quantisation value for each group from the assumed relationship between a number of bits per macroblock and Activity of the macroblock and the average Activity of the group, wherein the total number of bits available for all of the macroblocks in the coded picture is imposed by a predetermined bit rate of the video compression system.
2. The method as claimed in claim 1 , wherein estimating, in the processor, sizes of a picture comprises using, in the processor, a simplified coding process, including a simple transform, quantisation and variable length coding.
3. The method as claimed in claim 2 , comprising using, in the processor, an approximation of the H264 Context-adaptive binary arithmetic coding (CABAC) entropy encoding standard.
4. The method as claimed in claim 1 , wherein estimating, in the processor, sizes of a picture dependent on a plurality of quantisation values comprises estimating, in the processor, over a range of quantizer values selected to cover most of a quantisation space of a compression standard.
5. The method as claimed in claim 1 , wherein assuming, in the processor, a relationship comprises assuming, in the processor, a linear relationship in which constants may be determined by a least squares estimation.
6. The method as claimed in claim 1 , wherein dividing, in the processor, macroblocks into groups by Activity comprises determining, in the processor, a normalised sum of coded coefficients as a measure of the Activity of each macroblock.