Video decoder with reduced dynamic range transform with inverse transform shifting memory
A method for decoding video includes receiving quantized coefficients representative of a block of video representative of a plurality of pixels. The quantized coefficients are dequantized based upon a function of a remainder. The dequantized coefficients are inverse transformed to determine a decoded residue.
1. A method for decoding video comprising:
receiving quantized coefficients representative of a block of video representative of a plurality of pixels;
descaling the quantized coefficients based on a quantization parameter, a coefficient index, and a transform size to generate descaled coefficients;
applying an adjustment mechanism to the descaled coefficients to modify the descaled coefficients to generate modified descaled coefficients, wherein the adjustment mechanism is a variable based on the transform size;
clipping the modified descaled coefficients to a predetermined bit depth to generate clipped coefficients;
storing the clipped coefficients in a memory; and
inverse transforming the clipped coefficients to output a decoded residue, wherein the inverse transforming comprises:
one-dimension inverse transforming in a first direction the clipped coefficients to generate first direction inverse transformed coefficients;
prior to one-dimension inverse transforming the first direction inverse transformed coefficients in a second direction, shifting the first direction inverse transformed coefficients right to generate first shifted inverse transformed coefficients having a bit depth less than the first direction inverse transformed coefficients;
clipping the first shifted inverse transformed coefficients to the predetermined bit depth to generate second clipped coefficients; and
one-dimension inverse transforming the second clipped coefficients in the second direction to determine the decoded residue.
2. The method of claim 1 , further comprising:
selecting, from among multiple different quantities of bits by which to shift coefficients and based on a size of the block of video, a particular quantity of bits by which to shift the first inverse transformed coefficients; and
wherein shifting the first inverse transformed coefficients right includes:
shifting the first inverse transformed coefficients right by the particular quantity of bits.
3. The method of claim 1 , wherein the first direction is a horizontal direction.
4. The method of claim 1 , wherein the second direction is orthogonal to the first direction.
5. The method of claim 1 , wherein the first direction is a vertical direction.
6. The method of claim 1 , further comprising:
selecting, from among multiple different quantities of bits by which to shift coefficients, a particular quantity of bits by which to shift coefficients;
wherein shifting the first inverse transformed coefficients right to generate first shifted inverse transformed coefficients includes:
shifting the first inverse transformed coefficients right by the particular quantity of bits to generate first shifted inverse transformed coefficients.
7. The method of claim 6 , further comprising:
determining that the block of video is of a particular size; and
wherein selecting the particular quantity of bits by which to shift coefficients includes: selecting, from among multiple different quantities of bits by which to shift coefficients that each correspond to different sizes of blocks of video, a particular quantity of bits by which to shift coefficients that corresponds to the particular size of the block of video.
8. The method of claim 1 , wherein shifting the first direction inverse transformed coefficients right comprises shifting the first direction inverse transformed coefficients right by an amount that is based on a transform precision extension parameter.