METHODS AND DEVICES FOR RESIDUAL SCALING DEPENDING ON PREDICTION FOR VIDEO CODING

Abstract

Methods and devices are provided for reducing the decoding latency introduced by LMCS. In one method, during the decoding of an encoding unit (CU), a plurality of reconstructed luminance samples is selected from a first predetermined region adjacent to a second predetermined region where the CU is located, an average of the plurality of reconstructed luminance samples is calculated, and the average of the plurality of reconstructed luminance samples is used directly, without any clipping, in deriving a residual chrominance scaling factor to decode the CU.

Claims

NOVELTY OF THE INVENTION Having described the present invention, the following claims are considered novel and therefore claimed as priority: CLAIMS 1. A method for video coding, characterized in that it comprises: selecting a plurality of reconstructed luminance samples from a first predetermined region adjacent to a second predetermined region where an encoding unit (CU) is located; calculating an average of the plurality of reconstructed luminance samples; and deriving a residual chroma scale factor for decoding the CU using the average of the plurality of reconstructed luminance samples directly, without any clipping.

2. The method according to claim 1, characterized in that the average of the plurality of reconstructed luminance samples is the arithmetic mean of the plurality of reconstructed luminance samples.

3. The method according to claim 1, characterized in that deriving the residual chrominance scaling factor for decoding the CU using the average of the plurality of directly reconstructed luminance samples, without any clipping, comprises: identifying a segment index for the average in a predefined piece-level linear model; and deriving the residual chrominance scaling factor for decoding the CU based on the segment index.

4. The method according to claim 1, characterized in that the plurality of reconstructed luminance samples is generated by the following: generating luminance prediction samples and residual luminance samples; adding the residual luminance samples to the luminance prediction samples; and clipping the added luminance samples to a dynamic range of one encoding bit depth.

5. The method according to claim 1, characterized in that the plurality of reconstructed luminance samples is derived based on interpredicted samples with forward mapping.

6. The method according to claim 1, characterized in that the second predetermined region is a 64x64 region where the CU is located.

7. The method according to claim 6, characterized in that the first predetermined region comprises a 1x64 region directly above the second predetermined region and a 64x1 region directly to the left of the second predetermined region.

8. A computing device characterized in that it comprises: one or more processors; wherein one or more processors are configured to cause the computing device to perform acts comprising: selecting a plurality of reconstructed luminance samples from a first predetermined region adjacent to a second predetermined region where an encoding unit (CU) is located; calculating an average of the plurality of reconstructed luminance samples; and deriving a residual chromance scaling factor to decode the CU using the average of the plurality of reconstructed luminance samples directly, without any clipping.

9. The computing device according to claim 8, characterized in that the average of the plurality of reconstructed luminance samples is the arithmetic mean of the plurality of reconstructed luminance samples.

10. The computing device according to claim 8, characterized in that deriving the residual chrominance scaling factor for decoding the CU using the average of the plurality of directly reconstructed luminance samples, without any clipping, comprises: identifying a segment index for the average in a predefined piece-level linear model; and deriving the residual chrominance scaling factor for decoding the CU based on the segment index.

11. The computing device according to claim 8, characterized in that the plurality of reconstructed luminance samples is generated by: generating luminance prediction samples and residual luminance samples; adding the residual luminance samples to the luminance prediction samples; and clipping the aggregated luminance samples to a dynamic range of a coding bit depth.

12. The computing device according to claim 8, characterized in that the plurality of reconstructed luminance samples is derived based on interpredicted samples with forward mapping.

13. The computing device according to claim 8, characterized in that the second predetermined region is a 64x64 region where the CU is located.

14. The computing device according to claim 13, characterized in that the first predetermined region comprises a 1x64 region directly above the second predetermined region and a 64x1 region directly to the left of the second predetermined region.

Images