Systems and methods for scaling transform coefficient level values
A method for coding video data is disclosed. The method comprising: receiving a rectangular array of coefficient level values having a width and a height; and for each coefficient level value included in the rectangular array, generating a corresponding transform coefficient value, wherein generating a corresponding transform coefficient value includes scaling a coefficient level value based on a quantization parameter and the width and the height of the rectangular array.
1. A method of decoding a bitstream and deriving a transform coefficient value based on the bitstream by an electronic device, the method including:
receiving a level value from the bitstream;
deriving a shift value by using a first value specifying a bit depth of a color component and a second value calculated by dividing a sum of a base-2 logarithm of a transform block width and a base-2 logarithm of a transform block height by two,
deriving an intermediate value by adding a bit depth offset value to a level value scaled by a scaling factor, and
deriving the transform coefficient value by right shifting the intermediate value by the shift value for reconstructing the transform block.
2. The method of claim 1 , wherein the bit depth offset value is derived by left shifting the shift value by one bit.
3. The method of claim 1 , wherein the shift value is derived by subtracting five from a third value.
4. The device of claim 1 , wherein the scaling factor is derived by using a fourth value resulted from dividing a quantization parameter by 6.
5. The method of claim 1 , wherein the level value is derived by using the sum of the base-2 logarithm of the transform block width and the base-2 logarithm of the transform block height.
6. The method of claim 1 , wherein the transform block width and a transform block height are not equal.
7. A decoder for decoding coded data, the decoder comprising:
inverse quantization circuitry that derives a transform coefficient value by right shifting an intermediate value by a shift value; and
inverse transformation processing circuitry that derives an array of residual values by using the transform coefficient value,
wherein:
the shift value is derived by using a first value specifying a bit depth of a color component and a second value calculated by dividing a sum of a base-2 logarithm of a transform block width and a base-2 logarithm of a transform block height by two, and
the intermediate value is derived by adding a bit depth offset value to a level value scaled by a scaling factor.
8. The decoder of claim 7 , wherein the bit depth offset value is derived by left shifting the shift value by one bit.
9. The decoder of claim 7 , wherein the shift value is derived by subtracting five from a third value.
10. The decoder of claim 7 , wherein the scaling factor is derived by using a fourth value resulted from dividing a quantization parameter by 6.
11. The decoder of claim 7 , wherein the level value is derived by using the sum of the base-2 logarithm of the transform block width and the base-2 logarithm of the transform block height.
12. The decoder of claim 7 , wherein the transform block width and a transform block height are not equal.
13. An encoder for encoding video data, the encoder comprising:
inverse quantization circuitry that derives a transform coefficient value by right shifting an intermediate value by a shift value; and
inverse transformation processing circuitry that derives an array of residual values by using the transform coefficient value,
wherein:
the shift value is derived by using a first value specifying a bit depth of a color component and a second value calculated by dividing a sum of a base-2 logarithm of a transform block width and a base-2 logarithm of a transform block height by two, and
the intermediate value is derived by adding a bit depth offset value to a level value scaled by a scaling factor.
14. The encoder of claim 13 , wherein the bit depth offset value is derived by left shifting the shift value by one bit.
15. The encoder of claim 13 , wherein the shift value is derived by subtracting five from a third value.
16. The encoder of claim 13 , wherein the scaling factor is derived by using a fourth value resulted from dividing a quantization parameter by 6.
17. The encoder of claim 13 , wherein the level value is derived by using the sum of the base-2 logarithm of the transform block width and the base-2 logarithm of the transform block height.
18. The encoder of claim 13 , wherein the transform block width and a transform block height are not equal.