Motion vector prediction

Patent metadata

Field	Value
Publication number	US-9532066-B2
Application number	US-201113245668-A
Country	US
Kind code	B2
Filing date	Sep 26, 2011
Priority date	Jan 21, 2011
Publication date	Dec 27, 2016
Grant date	Dec 27, 2016

Abstract

A method of coding video data includes determining a candidate motion vector for each of one or more candidate portions of a video frame and determining a current motion vector for a current portion of a current frame. The current motion vector identifies a portion of a reference frame that at least partially matches the current portion of the current frame. The method also includes calculating a motion vector difference between the current motion vector and each of the candidate motion vectors, selecting one of the candidate motion vectors based on the calculated motion vector differences, signaling an index identifying the candidate portion having the selected one of the candidate motion vectors, and signaling the corresponding motion vector difference calculated with respect to the selected one of the candidate motion vectors.

First claim

What is claimed is:

1. A method of encoding video data, the method comprising:
determining a candidate motion vector for a motion vector prediction process for each of one or more candidate portions of a current video frame, wherein the one or more candidate portions are causal to a current portion of the current video frame;
determining a current motion vector for the current portion;
determining whether each of the candidate motion vectors and the current motion vector refer to the same reference video frame; and
performing the motion vector prediction process in the case that each of the candidate motion vectors and the current motion vector are determined to refer to the same reference video frame, whereby signaling of a reference index indicating the reference video frame is avoided by inferring the reference index from the candidate motion vectors of the one or more candidate portions at a video decoder, the motion vector prediction process comprising:
calculating a motion vector difference between the current motion vector and each of the candidate motion vectors;
selecting one of the candidate motion vectors based on the calculated motion vector differences;
not signaling the reference index;
signaling an index identifying the candidate portion having the selected one of the candidate motion vectors; and
signaling the motion vector difference calculated with respect to the selected one of the candidate motion vectors.

2. The method of claim 1 , wherein the one or more candidate portions are in the current video frame.

3. The method of claim 2 , wherein the candidate portions are adjacent to the current portion.

4. The method of claim 1 , wherein if the motion vector difference associated with the selected one of the candidate motion vectors corresponds to a motion vector difference of zero, the method further comprises signaling the corresponding motion vector difference determined with respect to the selected one of the candidate motion vectors comprises signaling a flag indicating the motion vector difference of zero.

5. The method of claim 1 , wherein selecting one of the candidate motion vectors based on the calculated motion vector differences comprises selecting the candidate motion vector with the smallest calculated motion vector difference.

6. A method of decoding video data, the method comprising:
performing a motion vector prediction process in the case that each of a plurality of candidate motion vectors and a current motion vector refers to the same reference video frame, whereby reception of a reference index indicating the reference video frame is avoided, the motion vector prediction process comprising:
receiving an index identifying a candidate portion of a current video frame, wherein the candidate portion is causal to a current portion of the current video frame;
receiving a motion vector difference;
retrieving a candidate motion vector associated with the candidate portion;
retrieving a motion prediction direction and the reference frame index associated with the candidate portion without receiving the reference frame index for the current portion;
determining the current motion vector for the current portion based on the received motion vector difference and the candidate motion vector; and
performing motion compensation on the current portion using the determined current motion vector, the retrieved motion prediction direction, and the retrieved reference frame index to produce predictive video data.

7. The method of claim 6 , further comprising:
entropy decoding video data for the current portion;
inverse quantizing and inverse transforming the entropy decoded video data to produce residual data; and
combining the predictive video data and the residual data to form decoded video data.

8. The method of claim 6 , wherein the received motion vector difference is a flag indicating a zero motion vector difference, and wherein determining the current motion vector for the current portion of the current video frame comprises copying the candidate motion vector associated with the candidate portion.

9. An apparatus comprising:
a memory configured to store a current video frame; and
a video encoder in communication with the memory, the video encoder configured to:
determine a candidate motion vector for a motion vector prediction process for each of one or more candidate portions of a current video frame, wherein the one or more candidate portions are causal to a current portion of the current video frame;
determine a current motion vector for the current portion;
determine whether each of the candidate motion vectors and the current motion vector refer to the same reference video frame;
perform the motion vector prediction process in the case that each of the candidate motion vectors and the current motion vector are determined to refer to the same reference video frame, whereby signaling of a reference index indicating the reference video frame is avoided by inferring the reference index from the candidate motion vectors of the one or more candidate portions at a video decoder, the video encoder further configured to:
calculate a motion vector difference between the current motion vector and each of the candidate motion vectors;
select one of the candidate motion vectors based on the calculated motion vector differences;
not signal the reference index;
signal an index identifying the candidate portion having the selected one of the candidate motion vectors; and
signal the motion vector difference calculated with respect to the selected one of the candidate motion vectors.

10. The apparatus of claim 9 , wherein the one or more candidate portions are in the current video frame.

11. The apparatus of claim 10 , wherein the candidate portions are adjacent to the current portion.

12. The apparatus of claim 9 , wherein if the motion vector difference associated with the selected one of the candidate motion vectors corresponds to a motion vector difference of zero, the video encoder is configured to signal the motion vector difference as a flag indicating the motion vector difference of zero.

13. The apparatus of claim 9 , wherein the video encoder is further configured to select the candidate motion vector with the smallest calculated motion vector difference.

14. An apparatus comprising:
a memory configured to store a current video frame; and
a video decoder in communication with the memory, the video decoder configured to perform a motion vector prediction process in the case that each of a plurality of candidate motion vectors and a current motion vector refers to the same reference video frame, whereby reception of a reference index indicating the reference video frame is avoided, the video decoder further configured to:
receive an index identifying a candidate portion of the current video frame, wherein the candidate portion is causal to a current portion of the current video frame;
receive a motion vector difference;
retrieve a candidate motion vector associated with the candidate portion;
retrieve a motion prediction direction and the reference frame index associated with the candidate portion without receiving the reference frame index for the current portion;
determine the current motion vector for the current portion based on the received motion vector difference and the candidate motion vector; and
perform motion compensation on the current portion using the determined current motion vector, the retrieved motion prediction direction, and the retrieved reference frame index to produce predictive video data.

15. The apparatus of claim 14

Assignees

• Chien Wei-Jung
• Chen Peisong
• Karczewicz Marta
• Qualcomm Inc

Inventors

• Chien Wei-Jung
• Chen Peisong
• Karczewicz Marta

Classifications

• H04N19/52

  by predictive encoding · CPC title

• H04N19/176

  the region being a block, e.g. a macroblock · CPC title

• H04N19/196

  being specially adapted for the computation of encoding parameters, e.g. by averaging previously computed encoding parameters (processing of motion vectors H04N19/513) · CPC title

• H04N19/463 Primary

  by compressing encoding parameters before transmission · CPC title

• H04N19/61 Primary

  in combination with predictive coding · CPC title