Method and apparatus for selecting a coding mode
View Patent ↗A method and apparatus for selecting a coding mode for a block of a current picture are disclosed. For example, the method selects a coding mode in accordance with a cost function, for coding the block, wherein the cost function comprises a coding distortion parameter and a number of coding bits parameter, wherein said coding distortion parameter is measured in accordance with at least one of: between a prediction residual and a reconstructed prediction residual, or between a transformed prediction residual and a dequantized transformed prediction residual, wherein the number of coding bits parameter is computed in accordance with at least one of: from a true number of compressed bits resulting from coding said block, directly from a plurality of bins, or directly from a plurality of quantized transform coefficients.
1. A method for selecting a coding mode for a block of a current picture, comprising:
receiving a block of a current picture;
subtracting a prediction block from the received block to calculate a prediction residual block;
performing a forward transform on the prediction residual block to calculate a transformed prediction residual block;
performing quantization on the transformed prediction residual block to calculate a quantized transformed prediction residual block;
performing a binarization process on the quantized transformed prediction residual block, resulting in a plurality of bins;
performing dequantization on the quantized transformed prediction residual block to calculate a dequantized transformed prediction residual block;
performing inverse transform on the dequantized transformed prediction residual block to calculate the reconstructed prediction residual block;
adding the reconstructed prediction residual block to the prediction block to calculate the reconstructed block;
computing a coding bits parameter from the plurality of bins;
measuring a coding distortion parameter between the received block and the reconstructed block; and
selecting a coding mode in accordance with a cost function for coding the received block, wherein the cost function comprises the coding distortion parameter and the coding bits parameter.
2. The method of claim 1 , wherein the cost function ƒ urther comprises a control parameter that is a function of a quantization parameter used for the received block.
3. The method of claim 1 , wherein said block is a macroblock.
4. The method of claim 1 , further comprising:
comparing said selected coding mode with a skip mode to determine a final coding mode for coding said block.
5. The method of claim 4 ,
wherein if a cost of said selected coding mode is lower than a cost of the skip mode, then said selected coding mode is used as said final coding mode for coding said block; or
wherein if a cost of said selected coding mode is higher than a cost of a skip mode, then said skip mode is used as said final coding mode for coding said block.