Method and device for decoding a color picture
The present disclosure generally relates to a device and a method of decoding a color picture from a bitstream. It comprises: obtaining an expanded range luminance as a first component by applying a nonlinear dynamic expansion function to a luminance component obtained from the bitstream, wherein said nonlinear dynamic expansion function is determined depending on an inverse of a dynamic reduction function that has been applied to an original luminance component obtained when encoding the color picture; obtaining a second component by calculating a square root of a difference between a value determined by the first component and another value computed as a linear combination of a multiplication product of two chrominance components and square values of the two chrominance components obtained from the bitstream, and obtaining at least one color component of the color picture to be decoded at least from said second component and said two chrominance components by determining said at least one color component as a linear combination of the second component and the two chrominance components.
1. A method of decoding a color picture, the method comprising:
obtaining a luminance component and two chrominance components of a standard dynamic range (SDR) picture;
obtaining two first intermediate chrominance components , {circumflex over (V)} r by multiplying the two chrominance components of the SDR picture by a factor;
obtaining a value {circumflex over (T)} as a combination of squares of the first intermediate chrominance components , {circumflex over (V)} r and a product of the first intermediate chrominance components , {circumflex over (V)} r ;
obtaining second intermediate chrominance components , by computing
= and ={circumflex over (V)} r when {circumflex over (T)}≤1
and as
= /√{square root over ({circumflex over ( T )})} and = {circumflex over (V)} r /√{square root over ({circumflex over (T)})} when {circumflex over (T)}> 1;
obtaining a value Ŝ as
{circumflex over ( S )}=√{square root over (1−{circumflex over ( T )})} when {circumflex over (T)} ≤1
and as
Ŝ=0 when {circumflex over (T)}>1;
obtaining first intermediate RGB values by multiplying a vector comprising the value Ŝ and the second intermediate chrominance components , by a matrix;
obtaining second intermediate RGB values by multiplying the first intermediate RGB values by a square root of a first intermediate luminance component; and
squaring each of the second intermediate RGB values to produce high dynamic range (HDR) samples.
2. The method of claim 1 , further comprising obtaining a first adjusted luminance component by adding a sum of the two chrominance components of the SDR picture to the luminance component of the SDR picture.
3. The method of claim 2 , wherein obtaining the first adjusted luminance component comprises clipping the sum of the two chrominance components of the SDR picture to a lower bound of zero.
4. The method of claim 2 , further comprising obtaining the square root of the first intermediate luminance component by applying a first function to the first adjusted luminance component.
5. The method of claim 4 , wherein the first function is a square root of a non-linear dynamic expansion function.
6. The method of claim 2 , wherein the factor is a function of the first adjusted luminance component.
7. The method of claim 2 , wherein the sum is a weighted sum.
8. The method of claim 1 , wherein the combination is a weighted combination.
9. An apparatus comprising:
at least one processor; and
at least one memory having stored instructions operative, when executed by the at least one processor, to cause the apparatus to decode a color picture by:
obtaining a luminance component and two chrominance components of a standard dynamic range (SDR) picture;
obtaining two first intermediate chrominance components , {circumflex over (V)} r by multiplying the two chrominance components of the SDR picture by a factor;
obtaining a value {circumflex over (T)} as a combination of squares of the first intermediate chrominance components , and a product of the first intermediate chrominance components , {circumflex over (V)} r ;
obtaining second intermediate chrominance components , by computing
= and ={circumflex over (V)} r when {circumflex over (T)}≤1
and as
= /√{square root over ({circumflex over ( T )})} and = {circumflex over (V)} r /√{square root over ({circumflex over ( T )})} when {circumflex over ( T )}>1;
obtaining a value Ŝ as
{circumflex over ( S )}=√{square root over (1−{circumflex over ( T )})} when {circumflex over (T)}> 1
and as
Ŝ=0 when {circumflex over (T)}>1;
obtaining first intermediate RGB values by multiplying a vector comprising the value Ŝ and the second intermediate chrominance components , by a matrix;
obtaining second intermediate RGB values by multiplying the first intermediate RGB values by a square root of a first intermediate luminance component; and
squaring each of the second intermediate RGB values to produce high dynamic range (HDR) samples.
10. The apparatus of claim 9 , wherein the instructions further cause the apparatus to obtain a first adjusted luminance component by adding a sum of the two chrominance components of the SDR picture to the luminance component of the SDR picture.
11. The apparatus of claim 10 , wherein obtaining the first adjusted luminance component comprises clipping the sum of the two chrominance components of the SDR picture to a lower bound of zero.
12. The apparatus of claim 10 , wherein the instructions further cause the apparatus to obtain the square root of the first intermediate luminance component by applying a first function to the first adjusted luminance component.
13. The apparatus of claim 12 , wherein the first function is a square root of a non-linear dynamic expansion function.
14. The apparatus of claim 10 , wherein the factor is a function of the first adjusted luminance component.
15. The apparatus of claim 10 , wherein the sum is a weighted sum.
16. The apparatus of claim 9 , wherein the combination is a weighted combination.
17. The apparatus of claim 9 , wherein a module or functional unit comprises the instructions.
18. A non-transitory computer-readable storage medium having stored instructions that, when executed by a processor, cause the processor to perform:
obtaining a luminance component and two chrominance components of a standard dynamic range (SDR) picture;
obtaining two first intermediate chrominance components , {circumflex over (V)} r by multiplying the two chrominance components of the SDR picture by a factor;
obtaining a value {circumflex over (T)} as a combination of squares of the first intermediate chrominance components , and a product of the first intermediate chrominance components , ;
obtaining second intermediate chrominance components , by computing
= and = {circumflex over (V)} r when {circumflex over (T)}≤ 1
and as
= /√{square root over ({circumflex over ( T )})} and = {circumflex over (V)} r /√{square root over ({circumflex over (T)})} when {circumflex over (T)}> 1;
obtaining a value Ŝ as
{circumflex over ( S )}=√{square root over (1−{circumflex over ( T )})} when {circumflex over (T)}≤ 1
and as
Ŝ=0 when {circumflex over (T)}>1;
obtaining first intermediate RGB values by multiplying a vector comprising the value Ŝ and the second intermediate chrominance components , by a matrix;
obtaining second intermediate RGB values by multiplying the first intermediate RGB values by a square root of a first intermediate luminance component; and
squaring each of the second intermediate RGB values to produce high dynamic range (HDR) samples.