High dynamic range adaptation operations at a video decoder

1. A method of decoding a bitstream, comprising:

receiving a bitstream compressed during encoding from a high-bit depth input in to a 10-bit 4:2:0 bitstream at a decoder;

decoding said bitstream with said decoder into color values and metadata items indicating information about adaptive post-processing operations to be performed by said decoder, said metadata including at least one of a color conversion, a transfer function, or a second color space conversion operation applied to the bitstream during encoding, said decoding including performing, with said decoder, one or more high dynamic range (HDR) adaptation operations on said color values based on said metadata items; and

reconstructing an HDR video from said color values by inverse quantization re-mapping the 10bit bitstream data, performing inverse perceptual normalization, and processing color components with an inverse color conversion, inverse transfer function, and inverse color conversion,

wherein said one or more HDR adaptation operations convert said color values into a format expected by said one or more fixed post-processing operations.

2. The method of claim 1 , wherein one of said fixed post-processing operations is an inverse color space conversion operation and one of said HDR adaptation operations is a color space adaptation operation that converts said color values from an input color space to an output color space expected by said inverse color space conversion operation.

3. The method of claim 2 , wherein said color space adaptation operation converts said color values based on a fixed lookup table or a fixed mapping function.

4. The method of claim 2 , wherein said color space adaptation operation converts said color values based on an adaptive lookup table or an adaptive mapping function determined from parameters in said metadata items.

5. The method of claim 1 , wherein said input color space is a YCbCr color space and said output color space is a perceptually quantized (PQ) YCbCr color space, and said decoder:

converts the Y luma component from said YCbCr color space to a PQ Y luma component for said PQ YCbCr color space using a luma conversion operation,

converts the Y luma component, the Cb chroma component, and the Cr chroma component from said YCbCr color space to an intermediate PQ blue component using a blue conversion operation,

converts the Y luma component, the Cb chroma component, and the Cr chroma component from said YCbCr color space to an intermediate PQ red component using a red conversion operation,

converts the PQ Y luma component, the intermediate PQ blue component, and the intermediate PQ red component into a PQ Cb chroma component using a Cb conversion operation, and

converts the PQ Y luma component, the intermediate PQ blue component, and the intermediate PQ red component into a PQ Cr chroma component using a Cr conversion operation,

wherein said luma conversion operation, said blue conversion operation, said red conversion operation, said Cb conversion operation, and said Cr conversion operation are defined with mapping functions or lookup tables.

6. The method of claim 5 , wherein one of said input color space and said output color space is a non-constant luminance color space and the other one of said input color space and said output color space is a constant luminance color space.

7. The method of claim 1 , wherein one of said fixed post-processing operations is an inverse color appearance transfer function and one of said HDR adaptation operations is a transfer function adaptation operation that converts said color values from being adjusted by an input transfer function to being adjusted by an output transfer function as expected by inverse color appearance transfer function.

8. The method of claim 7 , wherein one of said input transfer function and said output transfer function is a gamma function and the other one of said input transfer function and said output transfer function is a perceptual quantizer (PQ) function.

9. The method of claim 7 , wherein said transfer function adaptation operation converts said color values based on a fixed lookup table or a fixed mapping function.

10. The method of claim 9 , wherein said decoder has a fixed lookup table for each possible inverse color appearance transfer function that can be used by said decoder, and said decoder converts a color value by finding an output color value on one of the fixed lookup tables identified using said metadata items.

11. The method of claim 7 , wherein said transfer function adaptation operation converts said color values based on an adaptive lookup table or an adaptive mapping function determined from parameters in said metadata items.

12. The method of claim 11 , wherein said metadata items define a multiple piece-wise segment curve and said decoder generates said adaptive lookup table using parameters that define each segment of said multiple piece-wise segment curve.

13. The method of claim 12 , wherein said multiple piece-wise segment curve is a multiple piece-wise polynomial curve with an order of two.

14. The method of claim 1 , wherein one of said HDR adaptation operations is a bit depth adaptation operation that converts said color values from an input bit depth to an output bit depth when said fixed post-processing operations expects said color values at said output bit depth.

15. The method of claim 1 , wherein said metadata items indicate information about the encoder's pre-processing operations on a scene basis, a frame basis, or a frame region basis, such that said decoder changes said one or more HDR adaptation operations based on said metadata items for different scenes, frames, or frame regions.

16. A video decoder, comprising:

a data transmission interface configured to receive a bitstream compressed during encoding from a high-bit depth input in to a 10-bit 4:2:0 bitstream; and

a processor configured to:

decode said bitstream into color values and metadata items indicating information about adaptive post-processing operations to be performed by said processor, said metadata including at least one of a color conversion, a transfer function, or a second color space conversion operation applied to the bitstream during encoding, said decoding including

performing one or more high dynamic range (HDR) adaptation operations on said color values based on said metadata items; and

reconstruct an HDR video from said color values using inverse quantization re-mapping the 10 bit bitstream data, performing inverse perceptual normalization, and processing color components with an inverse color conversion, inverse transfer function, and inverse color conversion,

wherein said one or more HDR adaptation operations convert said color values into a format expected by said one or more fixed post-processing operations, and

wherein said one or more HDR adaptation operations convert the color space of said color values, convert the bit depth of said color values, and/or change a transfer function by which said color values are adjusted.

17. The video decoder of claim 16 , wherein said one or more HDR adaptation operations converts said color values based on fixed lookup tables or fixed mapping functions.

18. The video decoder of claim 16 , wherein said one or more HDR adaptation operations converts said color values based on adaptive lookup tables or adaptive mapping functions determined from parameters in said metadata items.

19. The video decoder of claim 16 , wherein said metadata items indicate information about the encoder's pre-processing operations on a scene basis, a frame basis, or a frame region basis, such that said decoder changes said one or more HDR adaptation operations based on said metadata items for different scenes, frames, or frame regions.