IP Library › Granted Patent US 9,998,757

Granted Patent B2

US 9,998,757 · App. 13/623,768 · Granted Jun 12, 2018

Reference picture signaling and decoded picture buffer management

Inventors: Ying Chen (San Diego, CA); Ye-Kui Wang (San Diego, CA)

Assignee: Velos Media, LLC

H04N19/58 H04N7/12 H04N19/105 H04N19/174 H04N19/176 H04N19/463 H04N19/50 H04N19/70

View Patent ↗

Loading inventors, assignments & file history…

Monitor This Case

Get email alerts when status or documents change.

Order Certified Copies

Most orders are placed with the USPTO same day — all within 24 business hours.

Order via The Patent Place →

Pre-filled with this patent's details

Quick Facts

Patent No.: US 9,998,757
App. No.: 13/623,768
Granted: Jun 12, 2018
Kind: B2

External Links

USPTO Patent Center↗ Full Patent PDF↗

Abstract

Techniques are described related to performing random access starting from a random access point picture that is not an instantaneous decoder refresh picture. Some techniques are also related to reducing the amount of information that is signaled for long-term reference pictures of a reference picture set. Additional techniques are also related to decoded picture buffer management, such as removing decoded pictures based on a temporal identification value.

Claims (138)

1. A method for coding video data, the method comprising:

coding, in a slice header of a slice of the video data, a full identifier value for a random access point (RAP) picture of the video data that is not an instantaneous decoder refresh (IDR) picture, wherein the full identifier value comprises a full picture order count (POC) value for the RAP picture;

determining a partial identifier value for a first non-RAP picture of the video data based on the full identifier value for the RAP picture;

determining a full identifier value for the first non-RAP picture, wherein the partial identifier value represents a portion of the full identifier value for the first non-RAP picture;

identifying the first non-RAP picture as a reference picture for a second non-RAP picture of the video data based on the full identifier value for the first non-RAP picture; and

inter-predicting the second non-RAP picture based on the first non-RAP picture.

2. The method of claim 1 , further comprising:

for non-RAP pictures following the RAP picture in decoding order, identifying reference pictures that can be used for inter-predicting the non-RAP pictures based on the full identifier value of the RAP picture.

3. The method of claim 1 , wherein coding the full identifier value comprises separately coding a most significant bit (MSB) portion of the full POC value for the RAP picture and a least signification bit (LSB) portion of the full POC value for the RAP picture.

4. The method of claim 1 , wherein the RAP picture comprises one of a clean random access (CRA) picture, a broken link access (BLA) picture, and a gradual decoding refresh (GDR) picture.

5. The method of claim 1 , further comprising:

coding a value of a flag, in the slice header for a current picture,

wherein when the value of the flag is a first value, no long-term reference picture can be used for inter-prediction of the current picture and no long-term reference picture can be used for inter-prediction of pictures following the current picture in decoding order, and

wherein when the value of the flag is a second value, one or more long-term reference pictures can be used for inter-prediction of the current picture and can be used for inter-prediction of the pictures following the current picture in decoding order.

6. The method of claim 1 , further comprising:

coding a value of a flag, in the slice header for a current picture,

wherein when the value of the flag is a first value, all long-term reference pictures specified in a parameter set can be used for inter-prediction of the current picture, and long-term reference pictures that are not specified in the parameter set and that are prior to the current picture in decoding order cannot be used for inter-prediction of the current picture and pictures following the current picture in decoding order, and

wherein when the value of the flag is a second value, not all of the long-term reference pictures specified in a parameter set can be used for inter-prediction of the current picture, and long-term reference pictures that are not specified in the parameter set and that are prior to the current picture in decoding order can be used for inter-prediction of the current picture and pictures following the current picture in decoding order.

7. The method of claim 6 , wherein, when the value of the flag is the first value, the method further comprising:

constructing a long-term reference picture subset of a reference picture set without coding indices into the parameter set that identify long-term reference pictures specified in the parameter set,

wherein constructing the long-term reference picture subset comprises including all of the long-term reference pictures specified in the parameter set in the reference picture subset.

8. The method of claim 1 , further comprising:

coding a value of a flag, in the slice header for a current picture,

wherein when the value of the flag is a first value, all long-term reference pictures specified in a parameter set can be used for inter-prediction of the current picture, and

wherein when the value of the flag is a second value, not all of the long-term reference pictures specified in the parameter set can be used for inter-prediction of the current picture.

9. The method of claim 1 , further comprising:

coding a value of a flag, in the slice header for a current picture,

wherein when the value of the flag is a first value, no long-term reference picture can be used for inter-prediction of the current picture, and all long-term reference pictures specified in a parameter set can be used for inter-prediction of pictures following the current picture in decoding order, and

wherein when the value of the flag is a second value, one or more of the long-term reference pictures can be used for inter-prediction of the current picture, and not all of the long-term reference pictures specified in the parameter set can be used for inter-prediction of pictures following the current picture in decoding order.

10. The method of claim 1 , further comprising:

deriving a reference picture set for a current picture;

determining a temporal identification value of a current picture; and

removing one or more pictures from a decoded picture buffer (DPB) having temporal identification values less than or equal to the temporal identification value of the current picture and that are not identified in the reference picture set,

wherein the current picture can be inter-predicted with reference to one or more pictures identified in the reference picture set whose temporal identification values are less than or equal to the temporal identification value of the current picture.

11. The method of claim 10 , further comprising:

determining that the one or more pictures are not needed for output,

wherein removing the one or more pictures comprises removing the one or more pictures from DPB having temporal identification values less than or equal to the temporal identification value of the current picture, that are not identified in the reference picture set, and that are not needed for output.

12. The method of claim 10 , wherein removing the one or more pictures comprises removing the one or more picture prior to inter-predicting the current picture.

13. A device for coding video data, the device comprising:

a decoded picture buffer (DPB); and

a video coder configured to:

code, in a slice header of a slice of the video data, a full identifier value for a random access point (RAP) picture of the video data that is not an instantaneous decoder refresh (IDR) picture, wherein the full identifier value comprises a full picture order count (POC) value for the RAP picture;

determine a partial identifier value for a first non-RAP picture of the video data based on the full identifier value for the RAP picture;

determine a full identifier value for the first non-RAP picture, wherein the partial identifier value represents a portion of full identifier value for the first non-RAP picture;

identify the first non-RAP picture as a reference picture for a second non-RAP picture of the video data based on the full identifier value for the first non-RAP picture; and

inter-predict the second non-RAP picture based on the first non-RAP picture.

14. The device of claim 13 , wherein, for non-RAP pictures following the RAP picture in decoding order, the video coder is configured to:

identify reference pictures that can be used for inter-predicting the non-RAP pictures based on the full identifier value of the RAP picture.

15. The device of claim 13 , wherein, to code the full identifier value, the video coder is configured to code a most significant bit (MSB) portion of the full POC value for the RAP picture and a least signification bit (LSB) portion of the full POC value for the RAP picture.

16. The device of claim 13 , wherein the RAP picture comprises one of a clean random access (CRA) picture, a broken link access (BLA) picture, and a gradual decoding refresh (GDR) picture.

17. The device of claim 13 , wherein the video coder is configured to:

code a value of a flag, in the slice header for a current picture,

18. The device of claim 13 , wherein the video coder is configured to:

code a value of a flag, in the slice header for a current picture,

19. The device of claim 18 , wherein, when the value of the flag is the first value, the video coder is configured to:

construct a long-term reference picture subset of a reference picture set without coding indices into the parameter set that identify long-term reference pictures specified in the parameter set,

wherein, to construct, the long-term reference picture subset, the video coder is configured to include all of the long-term reference pictures specified in the parameter set in the reference picture subset.

20. The device of claim 13 , wherein the video coder is configured to:

code a value of a flag, in the slice header for a current picture,

wherein when the value of the flag is a first value, all long-term reference pictures specified in a parameter set can be used for inter-prediction of the current picture, and

wherein when the value of the flag is a second value, not all of the long-term reference pictures specified in the parameter set can be used for inter-prediction of the current picture.

21. The device of claim 13 , wherein the video coder is configured to:

code a value of a flag, in the slice header for a current picture,

22. The device of claim 13 , wherein the video coder is configured to:

derive a reference picture set for a current picture;

determine a temporal identification value of a current picture; and

remove one or more pictures from the DPB having temporal identification values less than or equal to the temporal identification value of the current picture and that are not identified in the reference picture set,

23. The device of claim 22 , wherein the video coder is configured to:

determine that the one or more pictures are not needed for output,

wherein the video coder is configured to remove the one or more pictures from DPB having temporal identification values less than or equal to the temporal identification value of the current picture, that are not identified in the reference picture set, and that are not needed for output.

24. The device of claim 22 , wherein, to remove the one or more pictures, the video coder is configured to remove the one or more pictures prior to inter-predicting the current picture.

25. The device of claim 13 , wherein the video coder comprises a video decoder, and wherein to determine the full identifier value for the first non-RAP picture, the video decoder is configured to determine the full identifier value for the first non-RAP picture based on the partial identifier value for the first non-RAP picture.

26. The device of claim 13 , wherein the video coder comprises a video encoder.

27. The device of claim 13 , wherein the device comprises one of:

a wireless communication device;

a microprocessor; or

an integrated circuit.

28. A non-transitory computer-readable storage medium having stored thereon instructions that, when executed, cause a processor of a device for coding video data to:

code, in a slice header of a slice of the video data, a full identifier value for a random access point (RAP) picture that is not an instantaneous decoder refresh (IDR) picture, wherein the full identifier value comprises a full picture order count (POC) value for the RAP picture;

determine a partial identifier value for a first non-RAP picture of the video data based on the full identifier value for the RAP picture;

determine a full identifier value for the first non-RAP picture, wherein the partial identifier value represents a portion of the full identifier value for the non-RAP picture;

identify the first non-RAP picture as a reference picture for a second non-RAP picture of the video data based on the full identifier value for the first non-RAP picture; and

inter-predict the second non-RAP picture based on the first non-RAP picture.

29. The non-transitory computer-readable storage medium of claim 28 , further comprising instructions that cause the processor to:

code a value of a flag, in the slice header for a current picture,

30. The non-transitory computer-readable storage medium of claim 28 , further comprising instructions that cause the processor to:

code a value of a flag, in the slice header for a current picture,

31. The non-transitory computer-readable storage medium of claim 28 , further comprising instructions that cause the processor to:

code a value of a flag, in the slice header for a current picture,

wherein when the value of the flag is a first value, all long-term reference pictures specified in a parameter set can be used for inter-prediction of the current picture, and

wherein when the value of the flag is a second value, not all of the long-term reference pictures specified in the parameter set can be used for inter-prediction of the current picture.

32. The non-transitory computer-readable storage medium of claim 28 , further comprising instructions that cause the processor to:

code a value of a flag, in the slice header for a current picture,

33. The non-transitory computer-readable storage medium of claim 28 , further comprising instructions that cause the processor to:

derive a reference picture set for a current picture;

determine a temporal identification value of a current picture; and

remove one or more pictures from a decoded picture buffer (DPB) having temporal identification values less than or equal to the temporal identification value of the current picture and that are not identified in the reference picture set,

34. A device for coding video data, the device comprising:

means for coding, in a slice header of a slice of the video data, a full identifier value for a random access point (RAP) picture that is not an instantaneous decoder refresh (IDR) picture, wherein the full identifier value comprises a full picture order count (POC) value for the RAP picture;

means for determining a partial identifier value for a first non-RAP picture of the video data based on the full identifier value for the RAP picture;

means for determining a full identifier value for the first non-RAP picture, wherein the partial identifier value represents a portion of the full identifier value for the non-RAP picture;

means for identifying the first non-RAP picture as a reference picture for a second non-RAP picture of the video data based on the full identifier value for the first non-RAP picture; and

means for inter-predicting the second non-RAP picture based on the first non-RAP picture.

35. The device of claim 34 , further comprising:

means for coding a value of a flag, in the slice header for a current picture,

36. The device of claim 34 , further comprising:

means for coding a value of a flag, in the slice header for a current picture,

37. The device of claim 34 , further comprising:

means for coding a value of a flag, in the slice header for a current picture,

wherein when the value of the flag is a first value, all long-term reference pictures specified in a parameter set can be used for inter-prediction of the current picture, and

wherein when the value of the flag is a second value, not all of the long-term reference pictures specified in the parameter set can be used for inter-prediction of the current picture.

38. The device of claim 34 , further comprising:

means for coding a value of a flag, in the slice header for a current picture,

39. The device of claim 34 , further comprising:

means for deriving a reference picture set for a current picture;

means for determining a temporal identification value of a current picture; and

means for removing one or more pictures from a decoded picture buffer (DPB) having temporal identification values less than or equal to the temporal identification value of the current picture and that are not identified in the reference picture set,

Assignments (6)

RELEASE OF SECURITY INTEREST Recorded Jul 26, 2022

From: CORTLAND CAPITAL MARKET SERVICES LLC

To: VELOS MEDIA, LLC AND VELOS MEDIA MANAGEMENT, LLC

Reel/Frame 060923/0407 →

ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Sep 23, 2021

From: VELOS MEDIA, LLC

To: QUALCOMM INCORPORATED

Reel/Frame 057575/0431 →

RELEASE OF SECURITY INTEREST Recorded Sep 21, 2021

From: CORTLAND CAPITAL MARKET SERVICES LLC

To: VELOS MEDIA, LLC AND VELOS MEDIA MANAGEMENT, LLC