US 10,477,232 B2Granted

Search region determination for intra block copy in video coding

AssigneeQUALCOMM Incorporated

InventorsChao Pang, Joel Sole Rojals, Cheng-Teh Hsieh, Marta Karczewicz

Life of this patent

Estimating the term…

Loading patent…

Abstract

A video coder may determine a search region for coding a current block of video data using Intra Block Copy (Intra BC). In some examples, the video coder determines a central point for the search region, and determines the search region for the current block based on the central point and a defined size for the search region. The video coder stores reconstructed blocks of the video data from a current picture that includes the current block in a memory based on the determined search region. The video coder codes information from which to identify one of the reconstructed blocks within the search region, and codes the current block based on the identified one of the reconstructed blocks according to Intra BC.

Claims23

Dependent claims

A method of decoding video data, the method comprising:

determining that a current picture of the video data is an inter coded picture comprising both blocks that are spatially predicted by copying other blocks in the current picture and blocks that are temporally predicted from blocks in reference pictures that are different than the current picture;
determining that a current block of the current picture is coded using an intra block copy mode;
in response to the current block being coded using the intra block copy mode, determining a central point for a search region for decoding the current block using the intra block copy mode, wherein determining the central point for the search region comprises locating the central point for the search region using a block vector of a previously decoded block;
determining the search region for the current block based on the central point and a defined size of a rectangular region, wherein the central point for the search region is a central point of the rectangular region, and wherein the determined search region comprises less than a full available frame corresponding to the current picture;
storing reconstructed blocks of the current picture that are within the determined search region in a memory;
decoding information from which to identify one of the reconstructed blocks within the determined search region;
reconstructing the current block based on the identified one of the reconstructed blocks;
outputting a picture of decoded video data comprising the reconstructed current block;
determining that a second current picture of the video data is an intra coded picture comprising only blocks that are spatially predicted by copying other blocks in the second current picture;
determining that a second current block of the second current picture is coded using the intra block copy mode;
in response to the second current block being coded using the intra block copy mode, determining that a second search region for decoding the second current block of the video data using the intra block copy mode comprises a full available frame corresponding to the second current picture;
storing reconstructed blocks of the full available region of the second current picture in the memory, wherein the second current block belongs to the second current picture;
decoding second information from which to identify one of the reconstructed blocks within the second search region;
reconstructing the second current block based on the identified one of the reconstructed blocks within the second search region;
and
outputting a second picture of decoded video data comprising the reconstructed second current block.

13 more elements

A device for decoding video data, the device comprising:

a memory configured to store video data;
and
one or more processors configured to:
determine that a current picture of the video data is an inter coded picture comprising both blocks that are spatially predicted by copying other blocks in the current picture and blocks that are temporally predicted from blocks in reference pictures that are different than the current picture;
determine that a current block of the current picture is coded using an intra block copy mode;
in response to the current block being coded using the intra block copy mode, determine a central point for a search region for decoding the current block using the intra block copy mode, wherein to determine the central point for the search region, the one or more processors are configured to locate the central point for the search region using a block vector of a previously decoded block;
determine the search region for the current block based on the central point and a defined size of a rectangular region, wherein the central point for the search region is a central point of the rectangular region, and wherein the determined search region comprises less than a full available frame corresponding to the current picture;
store reconstructed blocks of the current picture that are within the determined search region in a memory;
decode information from which to identify one of the reconstructed blocks within the determined search region;
reconstruct the current block based on the identified one of the reconstructed blocks;
output a picture of decoded video data comprising the reconstructed current block;
determine that a second current picture of the video data is an intra coded picture comprising only blocks that are spatially predicted by copying other blocks in the second current picture;
determine that a second current block of a second current picture is coded using the intra block copy mode;
in response to the second current block being coded using the intra block copy mode, determine that a second search region for decoding the second current block of the video data using the intra block copy mode comprises a full available frame corresponding to the second current picture;
store reconstructed blocks of the full available region of the second current picture in the memory, wherein the second current block belongs to the second current picture;
decode second information from which to identify one of the reconstructed blocks within the second search region;
reconstruct the second current block based on the identified one of the reconstructed blocks within the second search region;
and
output a second picture of decoded video data comprising the reconstructed second current block.

16 more elements

A non-transitory computer-readable storage medium having instructions stored thereon that when executed cause one or more processors of a video decoder to:

determine that a current picture of the video data is an inter coded picture comprising both blocks that are spatially predicted by copying other blocks in the current picture and blocks that are temporally predicted from blocks in reference pictures that are different than the current picture;
determine that a current block of the current picture is coded using an intra block copy mode;
in response to the current block being coded using the intra block copy mode, determine a central point for a search region for decoding the current block using the intra block copy mode, wherein to determine the central point for the search region, the one or more processors are configured to locate the central point for the search region using a block vector of a previously decoded block;
determine the search region for the current block based on the central point and a defined size of a rectangular region, wherein the central point for the search region is a central point of the rectangular region, and wherein the determined search region comprises less than a full available frame corresponding to the current picture;
store only reconstructed blocks of the current picture that are within the determined search region in a memory;
decode information from which to identify one of the reconstructed blocks within the determined search region;
reconstruct the current block based on the identified one of the reconstructed blocks;
and
output a picture of decoded video data comprising the reconstructed current block;
determine that a second current picture of the video data is an intra coded picture comprising only blocks that are spatially predicted by copying other blocks in the second current picture;
determine that a second current block of a second current picture is coded using the intra block copy mode;
in response to the second current block being coded using the intra block copy mode, determine that a second search region for decoding the second current block of the video data using the intra block copy mode comprises a full available frame corresponding to the second current picture;
store reconstructed blocks of the full available region of the second current picture in the memory, wherein the second current block belongs to the second current picture;
decode second information from which to identify one of the reconstructed blocks within the second search region;
reconstruct the second current block based on the identified one of the reconstructed blocks within the second search region;
and
output a second picture of decoded video data comprising the reconstructed second current block.

14 more elements

Dependencies are inferred from the claim text.

Facts

Application: 14/663,161Filed 19 Mar 2015 · us provisional application
Priority: 21 Mar 2014
Published: US 2015/0271517 A124 Sep 2015
Granted: 12 Nov 2019
Term ends: Not estimatedThe source records don’t support an estimate.

Classification10 codes

H04N 19/423
Characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software…
H04N 19/52
By encoding: by predictive encoding
H04N 19/105
Selection of coding mode or of prediction mode: selection of the reference unit for prediction within a chosen coding or prediction mode, e.g.…
H04N 19/139
Motion inside a coding unit, e.g. average field, frame or block difference: analysis of motion vectors, e.g. their magnitude, direction, variance…