Inherited context initialization in CABAC

Abstract

Aspects of the present disclosure provide a method and apparatus for encoding / decoding video. In some examples, the apparatus for video decoding includes a receiving circuit and a processing circuit. The processing circuit determines the use of inherited context initialization in context-based adaptive binary arithmetic coding (CABAC) for decoding a current portion of a video sequence carried by a bitstream. The processing circuit decodes from the bitstream a syntax value indicating a decoded portion of the video sequence for context inheritance, and initializes a context model for CABAC based on a stored context state associated with the decoded portion. Furthermore, the processing circuit decodes the current portion from the bitstream according to the context model initialized based on the stored context state associated with the decoded portion.

Claims

1. A method of video processing performed by a decoder, The steps include determining the use of inherited context initialization in context-based adaptive binary arithmetic coding (CABAC) to decode the current portion of a video sequence carried by a bitstream, The steps include: decoding a syntax value from the bitstream that indicates the decoded portion of the video sequence for context inheritance; A step of initializing a context model for the CABAC based on the stored context state associated with the decoded portion, A method comprising the step of decoding the current portion from the bitstream according to the context model initialized based on the stored context state associated with the decoded portion.

2. The step of decoding the syntax value is: The steps include: decoding the syntax value that indicates the decoded picture for context inheritance, or The method according to claim 1, further comprising at least one of the steps of decoding the syntax value indicating the decoded slice for context inheritance.

3. The method according to claim 1, wherein the syntax value is decoded from at least one of a picture parameter set (PPS), a picture header, and a slice header.

4. The method according to claim 1, wherein the decoded portion is a reference used in the step of decoding the current portion.

5. A step of decoding the syntax value that represents at least a first decoded portion and a second decoded portion, A step of initializing the context model of the CABAC according to a weighted average of probability values ​​indicated by at least a first context state associated with the first decoded portion and a second context state associated with the second decoded portion. The method according to claim 1, further comprising:

6. The step of decoding the syntax value is: A step of decoding a first weight coefficient associated with the first decoded portion, The method according to claim 5, further comprising the step of decoding a second weighting coefficient associated with the second decoded portion.

7. The steps of calculating the weighted average include: applying the first weight coefficient to a first probability value indicated by the first context state associated with the first decoded portion, and applying the second weight coefficient to a second probability value indicated by the second context state associated with the second decoded portion. The method according to claim 6, further comprising:

8. Step 1: In response to a bitstream that does not carry the weight coefficient associated with the first decoded portion, decide to use a weight coefficient of zero or a default weight coefficient associated with the first decoded portion. The method according to claim 5, further comprising:

9. A step of decoding a Group of Pictures (GOP) level indicator from the bitstream, wherein the indicator indicates one or more pictures that are permitted to inherit the context; A step of prioritizing the decoding of the first picture of the one or more pictures before decoding a second picture that is not present in the one or more pictures. The method according to claim 1, further comprising:

10. The step of decoding the indicator at the GOP level is: The method according to claim 9, further comprising the step of decoding the indicator from a reference picturelist structure syntax.

11. An apparatus configured to perform the method described in any one of claims 1 to 10.

12. A computer program for causing a computer to perform the method described in any one of claims 1 to 10.

Images