US 8,477,853 B2Granted

Method and apparatus for encoding and/or decoding bit depth scalable video data using adaptive enhancement layer prediction

AssigneeThomson Licensing

InventorsYu Wen Wu, Yong Ying Gao

Life of this patent

Estimating the term…

Loading patent…

Abstract

A scalable video bitstream may have an H.264/AVC compatible base layer (BL) and a scalable enhancement layer (EL), where scalability refers to color bit depth. The SVC standard allows spatial inter-layer prediction, wherein a residual in the EL is generated which is then intra coded. Another spatial intra-coding mode for EL is pure intra coding (I_N×N). The invention discloses a new intra-coding mode and two new inter coding modes, particularly for bit depth scalability. The new intra coding mode uses encoding of the residual between upsampled reconstructed BL and original EL, using mode selection. Two possible modes are residual prediction from BL and additional intra-coding of this residual. The new inter coding modes use also prediction of EL from reconstructed BL. In a first inter coding mode, the residual is encoded using Motion Estimation based on this residual. In a second inter coding mode, the residual is encoded using upsampled motion information from the BL.

Claims6

Dependent claims

Method for encoding video data having a base layer and an enhancement layer, wherein the base layer has lower color resolution than the enhancement layer, the method comprising the steps of:

transforming and quantizing base layer data;
inverse transforming and inverse quantizing the transformed and quantized base layer data, wherein reconstructed base layer data are obtained;
upsampling the reconstructed base layer data, wherein the upsampling refers at least to bit depth and wherein a predicted version of the enhancement layer is obtained;
generating a residual between an original enhancement layer and the predicted version of the enhancement layer;
selecting for a case of an inter coded enhancement layer macroblock between at least two different inter coding modes and for a case of an intra coded enhancement layer macroblock between at least two different intra coding modes, wherein a first inter coding mode of the at least two different inter coding modes comprises using upsampled base layer motion information and a second inter coding mode of the at least two different inter coding modes comprises using motion information generated from said residual, and wherein at least one but not all of the intra coding modes comprises an additional step of intra coding of said residual using an Intra mode in which prediction from said residual is employed;
encoding the transformed and quantized base layer data;
encoding said residual using the selected mode and encoding an indication indicating said selected mode, wherein the encoding comprises a transform and a quantization step;
and
performing entropy coding on the encoded enhancement layer residual and said indication.

6 more elements

Method for decoding scalable video data having a base layer and an enhancement layer, wherein the base layer has less bit depth than the enhancement layer, comprising the steps of:

performing entropy decoding to obtain quantized and transformed enhancement layer information and base layer information and a decoding mode indication;
performing inverse quantization and inverse transformation on the base layer information and on the obtained quantized and transformed enhancement layer information to obtain a residual between a predicted enhancement layer and an original enhancement layer;
upsampling inverse quantized and inverse transformed base layer information, wherein a bit depth per value is increased and wherein the predicted enhancement layer is obtained;
and
reconstructing, from the predicted enhancement layer and the inverse quantized and inverse transformed enhancement layer information, reconstructed enhancement layer video information, wherein a decoding mode on a macroblock level according to said decoding mode indication is selected from
a first mode, wherein in a case of inter coded enhancement layer information the inverse quantized and inverse transformed enhancement layer information is decoded using motion information that is generated from the residual and extracted from the enhancement layer information;
a second mode, wherein in the case of inter coded enhancement layer information the inverse quantized and inverse transformed enhancement layer information is decoded using motion information that is extracted from the base layer information and then upsampled;
a third mode, wherein in a case of intra coded enhancement layer information prediction from said base layer is employed;
and
a fourth mode, wherein in the case of intra coded enhancement layer information the inverse quantized and inverse transformed enhancement layer information is intra decoded using an Intra mode in which prediction from said residual is employed.

7 more elements

Dependencies are inferred from the claim text.

Facts

Application: 12/448,153Filed 14 Dec 2006
Published: US 2010/0111167 A16 May 2010
Granted: 2 Jul 2013
Term ends: Not estimatedThe source records don’t support an estimate.

Classification4 codes

H04N 19/34
Using hierarchical techniques, e.g. scalability: scalability techniques involving progressive bit-plane based encoding of the enhancement layer,…
H04N 19/186
Characterised by the coding unit, i.e. the structural portion or semantic portion of the video signal being the object or the subject of the…: the…
H04N 19/30
Methods or arrangements for coding, decoding, compressing or decompressing digital video signals: using hierarchical techniques, e.g. scalability
H04N
Pictorial communication