Method and video decoder for decoding scalable video stream using inter-layer racing scheme

Patent metadata

Field	Value
Publication number	US-9838701-B2
Application number	US-201114235801-A
Country	US
Kind code	B2
Filing date	Aug 3, 2011
Priority date	Aug 3, 2011
Publication date	Dec 5, 2017
Grant date	Dec 5, 2017

Abstract

One exemplary method for decoding a scalable video stream, including a base layer frame and at least an enhancement layer frame corresponding to the base layer frame, has the following steps: decoding the base layer frame; and before the base layer frame is fully decoded, decoding the enhancement layer frame. Another exemplary method for decoding a scalable video stream, including a base layer frame and at least an enhancement layer frame corresponding to the base layer frame, has the following steps: decoding the enhancement layer frame, and decoding the base layer frame; wherein a start point of decoding the enhancement layer frame is earlier than a start point of decoding the base layer frame.

First claim

The invention claimed is:

1. A method for decoding a scalable video stream including a base layer frame and at least an enhancement layer frame corresponding to the base layer frame, comprising:
decoding the enhancement layer frame,
decoding the base layer frame, comprising:
decoding a portion of the base layer frame to generate a first partial decoding result;

when information derived from decoding the enhancement layer frame indicates that an inter-layer prediction between the base layer frame and the enhancement layer frame is not needed by decoding of a portion of the enhancement layer frame, not storing the first partial decoding result into a first data storage device; and
when the information derived from decoding the enhancement layer frame indicates that the inter-layer prediction is needed by decoding of the portion of the enhancement layer frame, storing the first partial decoding result into the first data storage device;
wherein a start point of decoding the enhancement layer frame is earlier than a start point of decoding the base layer frame.

2. The method of claim 1 , wherein the step of decoding the enhancement layer frame comprises:
parsing the enhancement layer frame to obtain prediction mode information;

the method further comprises:
checking if the inter-layer prediction is needed by referring to the prediction mode information, and accordingly generating a checking result;
wherein the first partial decoding result is selectively stored into the first data storage device according to the checking result.

3. The method of claim 2 , wherein the step of decoding the enhancement layer frame further comprises:
decoding the portion of the enhancement layer frame to generate a second partial decoding result; and
storing the second partial decoding result into a second data storage device;
when the checking result indicates that inter-layer prediction is needed by decoding of the portion of the enhancement layer frame, outputting a final decoding result of the portion of the enhancement layer frame by refining the second partial decoding result read from the second data storage device according to the first partial decoding result read from the first data storage device; and
when the checking result indicates that the inter-layer prediction is not needed by decoding of the portion of the enhancement layer frame, outputting the final decoding result of the portion of the enhancement layer frame by directly reading the second partial decoding result from the second data storage device.

4. The method of claim 3 , wherein a data size of the second partial decoding result is smaller than a data size of the first partial decoding result.

5. The method of claim 4 , wherein the data size of the first partial decoding result is equal to a data size of a plurality of rows of macroblocks.

6. The method of claim 2 , wherein the prediction mode information includes first macroblock-level syntax elements, second macroblock-level syntax elements, and third macroblock-level syntax elements of a plurality of macroblocks; and the step of checking if the inter-layer prediction is needed comprises:
generating the checking result indicating that the inter-layer prediction is needed when the first macroblock-level syntax elements indicate that an inter-layer intra prediction is used, the second macroblock-level syntax elements indicate that an inter-layer residual prediction is used, or the third macroblock-level syntax element indicate that an inter-layer motion prediction is used; and
generating the checking result indicating that the inter-layer prediction is not needed when the first macroblock-level syntax elements indicate that the inter-layer intra prediction is not used, the second macroblock-level syntax elements indicate that the inter-layer residual prediction is not used, and the third macroblock-level syntax elements indicate that the inter-layer motion prediction is not used.

7. A video decoder for decoding a scalable video stream including a base layer frame and at least an enhancement layer frame corresponding to the base layer frame, comprising:
a first data storage device;
an enhancement layer decoding circuit, arranged for decoding the enhancement layer frame; and
a base layer decoding circuit, arranged for decoding the base layer frame, comprising:
decoding a portion of the base layer frame to generate a first partial decoding result;
wherein when information derived from decoding the enhancement layer frame indicates that an inter-layer prediction between the base layer frame and the enhancement layer frame is not needed by decoding of a portion of the enhancement layer frame, the base layer decoding circuit does not store the first partial decoding result into the first data storage device; and when the information derived from decoding the enhancement layer frame indicates that the inter-layer prediction is needed by decoding of the portion of the enhancement layer frame, the base layer decoding circuit stores the first partial decoding result into the first data storage device;

wherein the enhancement layer decoding circuit starts decoding the enhancement layer frame before the base layer decoding circuit starts decoding the base layer frame.

8. The video decoder of claim 7 , wherein the enhancement layer decoding circuit is arranged for parsing the enhancement layer frame to obtain prediction mode information; and the video decoder further comprises:
a prediction mode information storage device, arranged for buffering the prediction mode information; and
a decision circuit, coupled to the prediction mode information storage device, the decision circuit arranged for checking if the inter-layer prediction is needed by referring to the prediction mode information, and accordingly generating a checking result; wherein the first partial decoding result is selectively stored into the first data storage device according to the checking result.

9. The video decoder of claim 8 , further comprising: a second data storage device, coupled to the enhancement layer decoding circuit; wherein the enhancement layer decoding circuit is arranged for decoding the portion of the enhancement layer frame to generate a second partial decoding result, and storing the second partial decoding result into the second data storage device; when the checking result indicates that inter-layer prediction is needed by decoding of the portion of the enhancement layer frame, outputting a final decoding result of the portion of the enhancement layer frame by refining the second partial decoding result read from the second data storage device according to the first partial decoding result read from the first data storage device; and when the checking result indicates that the inter-layer prediction is not needed by decoding of the portion of the enhancement layer frame, outputting the final decoding result of the portion of the enhancement layer frame by directly reading the second partial decoding result from the second data storage device.

10. The video decoder of claim 9 , wherein a data size of the second partial decoding result is smaller than a data size of the first partial decoding result.

11. The video decoder of claim 10 , wherein the data size of the first partial decoding result is equal to a data size of a plurality of rows of macroblocks.

12. The video decoder of claim 8 , wherein the prediction mode information includes first macroblock-level syntax elements, second macroblock-level syntax elements, and third macroblock-level syntax elements; and the decision circuit is arranged for generating the checking result indicating that the inter-layer prediction

Assignees

• Liu Tsu-Ming
• Chang Yung-Chang
• Ju Chi-Cheng
• Mediatek Inc

Inventors

• Liu Tsu-Ming
• Chang Yung-Chang
• Ju Chi-Cheng

Classifications

• H04N19/33

  in the spatial domain · CPC title

• H04N19/423

  characterised by memory arrangements (H04N19/433 takes precedence) · CPC title

• H04N19/426

  using memory downsizing methods · CPC title

• H04N19/31 Primary

  in the temporal domain · CPC title

• H04N19/44 Primary

  Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder · CPC title