Buffer compression for motion vector competition

Patent metadata

Field	Value
Publication number	US-9749657-B2
Application number	US-201113101451-A
Country	US
Kind code	B2
Filing date	May 5, 2011
Priority date	Jan 21, 2011
Publication date	Aug 29, 2017
Grant date	Aug 29, 2017

Abstract

Deriving a motion vector of a current block, in a current frame, including: generating a motion vector predictor candidate list including a set of first motion vectors and a second motion vector, wherein a first motion vector of the set belongs to a block adjacent to the current block in the current frame, and the second motion vector is a temporal motion vector of a pixel at a modified position (x′, y′) in a previous frame, the modified position (x′, y′) being derived by modifying a top-left position (x, y) of the current block. The method also includes deriving the modified position (x′, y′) using arithmetic shift operations; setting the second motion vector equal to a motion vector at the modified position (x′, y′) of the pixel in the previous frame; and deriving the motion vector of the current block using a motion vector predictor and a difference motion vector.

First claim

I claim:

1. A method for deriving a motion vector of a current block, in a current frame, the method comprising:
generating a motion vector predictor candidate list, wherein the motion vector predictor candidate list comprises a set of first motion vectors and a second motion vector;
wherein a first motion vector, in the set of first motion vectors, belongs to at least one adjacent block, which is adjacent to the current block in the current frame;
wherein the second motion vector is a temporal motion vector belonging to a pixel at a modified position (x′, y′) in a previous frame, the modified position (x′, y′) derived by modifying a top-left position (x, y) of the current block;
deriving the modified position (x′, y′) by: (i) setting a first intermediate position as (x+N/2, y+M/2) where N is a horizontal size of the current block and M is a vertical size of the current block, (ii) setting a second intermediate position by performing an arithmetic right shift operation to the first intermediate position, and thereafter (iii) setting the position (x′, y′) by performing an arithmetic left shift operation to the second intermediate position;
setting the second motion vector equal to a motion vector at the modified position (x′, y′) of the pixel in the previous frame; and
deriving the motion vector of the current block by using a motion vector predictor, which is a motion vector from the motion vector predictor candidate list, and a difference motion vector, which is the difference between the motion vector predictor and an actual motion vector.

2. The method of claim 1 , wherein the list is created by adding available motion vectors.

3. The method of claim 1 , wherein the motion vectors in the adjacent block includes a motion vector from a block located above a current block, a motion vector from a block located to the left of the current block, a motion vector from a block located above and to the left of the current block, a motion vector from a block located above and to the right of the current block, and a motion vector from a block located below and to the left of the current block.

4. The method of claim 1 , wherein motion vectors corresponding to the co-located block are assumed to be 0 if the co-located block is an intra-prediction block in the previous frame.

5. The method of claim 1 wherein the second motion vector is only used if the pixel at a particular location in the previous frame is coded using an inter-prediction technique.

Assignees

• Segall Christopher A
• Sharp Kk

Inventors

• Segall Christopher A

Classifications

• H04N19/61 Primary

  in combination with predictive coding · CPC title

• H04N19/513

  Processing of motion vectors · CPC title