Block-based advanced residual prediction for 3D video coding

Patent metadata

Field	Value
Publication number	US-9967592-B2
Application number	US-201514592633-A
Country	US
Kind code	B2
Filing date	Jan 8, 2015
Priority date	Jan 11, 2014
Publication date	May 8, 2018
Grant date	May 8, 2018

Abstract

Techniques for advanced residual prediction (ARP) in video coding may include receiving a first encoded block of video data in a first access unit, wherein the first encoded block of video data was encoded using advanced residual prediction and bi-directional prediction, determining temporal motion information for a first prediction direction of the first encoded block of video data, and identifying reference blocks for a second prediction direction, different than the first prediction direction, using the temporal motion information determined for the first prediction direction, wherein the reference blocks are in a second access unit.

First claim

What is claimed is:

1. A method for decoding video data, the method comprising:
receiving a first encoded block of video data in a first access unit of a first view, wherein the first encoded block of video data was encoded using advanced residual prediction and bi-directional prediction, the bi-directional prediction including temporal prediction for a first prediction direction and inter-view prediction for a second prediction direction, and wherein the advanced residual prediction includes performing a prediction of a residual of the first encoded block of video data relative to a residual of a residual predictor block;
determining first temporal motion information for the first prediction direction for the first encoded block of video data;
identifying a first temporal reference block for the first prediction direction using the first temporal motion information for the first prediction direction, wherein the first temporal reference block is in the first view and in a second access unit, the second access unit different than the first access unit;
deriving a disparity vector for the first prediction direction for the first encoded block of video data;
identifying a first inter-view reference block using the disparity vector for the first prediction direction, wherein the first inter-view reference block is in a second view, the second view being different than the first view;
identifying a second inter-view reference block using the first inter-view reference block and the first temporal motion information for the first prediction direction, the second inter-view reference block being in the second view;
determining a first residual predictor block from the first inter-view reference block and the second inter-view reference block;
determining first disparity motion information for the second prediction direction for the first encoded block of video data;
identifying a second temporal reference block for the second prediction direction using the first temporal motion information for the first prediction direction, wherein the second temporal reference block is the same as the first temporal reference block;
identifying a third inter-view reference block using the first disparity motion information for the second prediction direction, wherein the third inter-view reference block is in the second view;
identifying a fourth inter-view reference block using the third inter-view reference block and the first temporal motion information for the first prediction direction, the second inter-view reference block being in the second view;
determining a second residual predictor block from the third inter-view reference block and the fourth inter-view reference block;
and
performing advanced residual prediction on the first encoded block of video data using the first temporal reference block, the second temporal reference block, the first residual predictor block and the second residual predictor block.

2. The method of claim 1 , the method further comprising:
decoding the first encoded block of video data using one of block-level advanced residual prediction or prediction unit-level advanced residual prediction to produce residual video data; and
decoding the residual data using bi-directional prediction with the first temporal reference block for the first prediction direction and the second temporal reference block for the second prediction direction to produce a decoded block of video data.

3. An apparatus configured to decode video data, the apparatus comprising:
a video data memory configured to store a first encoded block of video data in a first access unit of a first view, wherein the first encoded block of video data was encoded using advanced residual prediction and bi-directional prediction, the bi-directional prediction including temporal prediction for a first prediction direction and inter-view prediction for a second prediction direction and wherein the advanced residual prediction includes performing a prediction of a residual of the first encoded block of video data relative to a residual of a residual predictor block; and
one or more processors in communication with the video data memory and configured to:
determine first temporal motion information for the first prediction direction for the first encoded block of video data;
identify a first temporal reference block for the first prediction direction using the first temporal motion information for the first prediction direction, wherein the first temporal reference block is in the first view and in a second access unit, the second access unit different than the first access unit;
derive a disparity vector for the first prediction direction for the first encoded block of video data;
identify a first inter-view reference block using the disparity vector for the first prediction direction, wherein the first inter-view reference block is in a second view, the second view being different than the first view;
identify a second inter-view reference block using the first inter-view reference block and the first temporal motion information for the first prediction direction, the second inter-view reference block being in the second view;
determine a first residual predictor block from the first inter-view reference block and the second inter-view reference block;
determine first disparity motion information for the second prediction direction for the first encoded block of video data;
identify a second temporal reference block for the second prediction direction using the first temporal motion information for the first prediction direction, wherein the second temporal reference block is the same as the first temporal reference block;
identify a third inter-view reference block using the first disparity motion information for the second prediction direction, wherein the third inter-view reference block is in the second view;
identify a fourth inter-view reference block using the third inter-view reference block and the first temporal motion information for the first prediction direction, the second inter-view reference block being in the second view;
determine a second residual predictor block from the third inter-view reference block and the fourth inter-view reference block;
and
perform advanced residual prediction on the first encoded block of video data using the first temporal reference block, the second temporal reference block, the first residual predictor block and the second residual predictor block.

4. The apparatus of claim 3 , wherein the one or more processors are further configured to:
decode the first encoded block of video data using one of block-level advanced residual prediction or prediction unit-level advanced residual prediction to produce residual video data; and
decode the residual data using bi-directional prediction with the first temporal reference block for the first prediction direction and the second temporal reference block for the second prediction direction to produce a decoded block of video data.

5. The apparatus of claim 4 , further comprising:
a display configured to display the decoded block of video data.

6. The apparatus of claim 3 , wherein the video data memory and the one or more processors are housed within one of a desktop computer, a notebook computer, a laptop computer, a set-top box, a telephone handset, a smart phone, a smart pad, a tablet computer, a television, a camera, a digital media player, a video gaming console, or a video streaming device.

7. An apparatus configured to decode video data, the apparatus comprising:
means for receiving a first encoded block of video data in a first access unit of a first view, wherein the first encoded block of video data was encoded using advanced residual prediction and bi-directional prediction,

Assignees

• Qualcomm Inc

Inventors

• Zhang Li
• Chen Ying

Classifications

• H04N19/597

  specially adapted for multi-view video sequence encoding · CPC title

• H04N19/577 Primary

  Motion compensation with bidirectional frame interpolation, i.e. using B-pictures · CPC title

• H04N19/52

  by predictive encoding · CPC title

• H04N19/105

  Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction · CPC title

• H04N19/176

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