Method and apparatus for fine-grained motion boundary processing

• US 9,813,730 B2
• Filed: 11/28/2014
• Issued: 11/07/2017
• Est. Priority Date: 12/06/2013
• Status: Active Grant

• Current Assignee: Xueshan Technologies Inc. (Wi-LAN Inc.)
• Original Assignee: MediaTek, Inc.
• Inventors: Hsu, Chih-Wei, Chen, Ching-Yeh
• Primary Examiner(s): Rahman, Mohammad J
• Application Number:
  US14/555,901
• Publication Number:
  US 20150163509A1
• Time in Patent Office: 1,075 Days
• Field of Search: None
• CPC Class Codes: H04N 19/567 Motion estimation based on rate distortion criteria
• CPC Class Codes: H04N 19/583 Motion compensation with overlapping blocks

Abstract

A method and apparatus for deriving fine-grained motion compensated prediction of boundary pixels in a video coding system are disclosed. Embodiments of the present invention determine one or more neighboring coding units (CUs) adjacent to a current coding unit (CU). For each neighboring CU, motion-compensated prediction is derived for each neighboring CU using the MV of the neighboring CU. The pre-generated predictors at a bottom side or a right side of each neighboring CUs are derived and stored on a smallest CU (SCU) basis. The pre-generated predictors and the motion compensated predictor for a current boundary pixel are combined using weighting factors to form a final predictor for the current pixel.

Claims

1. A method of fine-grained motion compensated prediction for boundary pixels in a video coding system, the method comprising:

determining one or more neighboring coding units (CUs) associated with a current coding unit (CU), wherein each of said one or more neighboring CUs is associated with a neighboring motion vector (MV);

performing motion-compensated prediction using the neighboring MV for each of said one or more neighboring CUs to derive pre-generated predictors, wherein the pre-generated predictors correspond to one or more boundary lines or columns in a boundary region of the current CU;

storing the pre-generated predictors;

receiving input data associated with the current CU having a current MV;

generating a first predictor for a current boundary pixel in the boundary region by applying motion compensation based on the current MV, wherein pixels at boundaries of the current CU utilize the current MV and at least one MV from at least one of;

an upper side MV and a left side MV to form a weighted sum of motion prediction when performing motion compensation;

generating a current boundary pixel predictor for the current boundary pixel using a weighted sum of the first predictor and one or more corresponding pre-generated predictors according to weighting factors; and

applying encoding or decoding to the current CU using prediction data including the current boundary pixel predictor,

wherein said pre-generated predictors are at a bottom side or a right side of each of said one or more neighboring CUs on a smallest CU (SCU) basis, and wherein said pre-generated predictors are stored on a SCU basis.

2. The method of claim 1, wherein a number of said one or more boundary lines or columns in the boundary region of the current CU is pre-defined or adaptively determined based on CU size or PU size.

3. The method of claim 1, wherein when the current CU comprises a luma component and at least one chroma component, the current boundary pixel predictor for the current boundary pixel also comprises the luma component and said at least one chroma component.

4. The method of claim 1, wherein a local line buffer is used to store the pre-generated predictors corresponding to said one or more boundary lines or columns in the boundary region of the current CU.

5. The method of claim 4, wherein size of the local line buffer is pre-defined.

6. The method of claim 4, wherein size of the local line buffer is proportional to frame width.

7. The method of claim 1, wherein the first predictor for a current boundary pixel in the boundary region is generated after said performing motion-compensated prediction using the neighboring MV for each of said one or more neighboring CUs.

8. The method of claim 1, wherein said one or more corresponding pre-generated predictors are collocated with the current boundary pixel.

17. The method of claim 1, wherein weighting factors for the weighted sum of motion prediction are adaptively determined based on a distance between a current boundary pixel and one of a left CU boundary and an upper CU boundary.

9. An apparatus of fine-grained motion compensated prediction for boundary pixels in a video coding system, the apparatus comprising one or more electronic circuits configured to:

determine one or more neighboring coding units (CUs) associated with a current coding unit (CU), wherein each of said one or more neighboring CUs is associated with a neighboring motion vector (MV);

performing motion-compensated prediction using the neighboring MV for each of said one or more neighboring CUs to derive pre-generated predictors, wherein the pre-generated predictors correspond to one or more boundary lines or columns in a boundary region of the current CU;

store the pre-generated predictors;

receive input data associated with the current CU having a current MV;

generate a first predictor for a current boundary pixel in the boundary region by applying motion compensation based on the current MV, wherein pixels at boundaries of the current CU utilize the current MV and at least one MV from at least one of;

an upper side MV and a left side MV to form a weighted sum of motion prediction when performing motion compensation;

generate a current boundary pixel predictor for the current boundary pixel using a weighted sum of the first predictor and one or more corresponding pre-generated predictors according to weighting factors; and

apply encoding or decoding to the current CU using prediction data including the current boundary pixel predictor,

wherein said pre-generated predictors are at a bottom side or a right side of each of said one or more neighboring CUs on a smallest CU (SCU) basis, and wherein said pre-generated predictors are stored on a SCU basis.

10. The apparatus of claim 9, wherein a number of said one or more boundary lines or columns in the boundary region of the current CU is pre-defined or adaptively determined based on CU size or PU size.

11. The apparatus of claim 9, wherein when the current CU comprises a luma component and at least one chroma component, the current boundary pixel predictor for the current boundary pixel also comprises the luma component and said at least one chroma component.

12. The apparatus of claim 9 further comprising a local line buffer to store the pre-generated predictors corresponding to said one or more boundary lines or columns in the boundary region of the current CU.

13. The apparatus of claim 12, wherein size of the local line buffer is pre-defined.

14. The apparatus of claim 12, wherein size of the local line buffer is proportional to frame width.

15. The apparatus of claim 9, wherein the first predictor for a current boundary pixel in the boundary region is generated after the motion-compensated prediction is performed using the neighboring MV for each of said one or more neighboring CUs.

16. The apparatus of claim 9, wherein said one or more corresponding pre-generated predictors are collocated with the current boundary pixel.

18. The apparatus of claim 9, wherein weighting factors for the weighted sum of motion prediction are adaptively determined based on a distance between a current boundary pixel and one of a left CU boundary and an upper CU boundary.