Multi-view signal codec with reusing coding parameters
Findings are described according to which a higher compression rate or better rate/distortion ratio may be achieved by adopting or predicting second coding parameters used for encoding a second view of the multi-view signal from first coding parameters used in encoding a first view of the multi-view signal. In other words, redundancies between views of a multi-view signal may not be restricted to the views themselves, such as the video information thereof, but that the coding parameters in parallelly encoding these views show similarities which may be exploited in order to further improve the coding rate.
1 . A decoder to generate a multi-view video, the multi-view video including a plurality of views, the decoder comprising:
a view decoder configured, using a processor, to:
extract, from a data stream, first information associated with a first coding block in a first view of the plurality of views of the multi-view video signal, wherein the first information is indicative of whether a first motion parameter of the first coding block is to:
(a) be read from the data stream independent of another coding parameter, or
(b) reuse one or more coding parameters of a second coding block located in a second view of the plurality of views of the multi-view video signal;
if the first information indicates that the first motion parameter of the first coding block is to be read from the data stream independent of another coding parameter, extract the first motion parameter from the data stream;
if the first information indicates that the first motion parameter of the first coding block is to reuse one or more coding parameters of the second coding block located in the second view,
obtain the one or more coding parameters including a second motion parameter of the second coding block, and
adopting the second motion parameter as the first motion parameter;
generate a prediction of the first coding block of the first view based at least on the extracted or adopted first motion parameter;
obtain from the data stream, residual data associated with the first coding block of the first view of the multi-view video signal; and
reconstruct the first coding block using the prediction of the first coding block and the residual data to produce a part of a video frame in the first view of the multi-view video.
2 . The decoder of claim 1 , wherein if the first information indicates that the first motion parameter of the first coding block is to reuse one or more coding parameters of the second coding block located in the second view, the view decoder is configured to extract from the data stream, second information identifying the second view from among the plurality of views of the multi-view video signal.
3 . The decoder of claim 2 , wherein the second information includes one or more reference frame indexes of the second view.
4 . The decoder of claim 1 , wherein the prediction of the first coding block is generated using one of intra-prediction and inter-prediction.
5 . The decoder of claim 4 , wherein the inter-prediction is inter-frame prediction.
6 . The decoder of claim 1 , wherein the first coding block is reconstructed using a different special resolution than the second coding block.
7 . The decoder of claim 1 , wherein the first coding block includes one of video data and a depth map.
8 . The decoder of claim 7 , wherein, if the first information indicates that the first motion parameter of the first coding block is to reuse the one or more coding parameters of the second coding block, the video data of the first coding block is reconstructed based on a first subset of the one or more coding parameters of the second coding block.
9 . The decoder of claim 7 , wherein, if the first information indicates that the first motion parameter of the first coding block is to reuse the one or more coding parameters of the second coding block, the depth map of the first coding block is reconstructed based on a second subset of the one or more coding parameters of the second coding block.
10 . An encoder for generating an encoded data stream of a multi-view video, the multi-view video including a plurality of views, the encoder comprising:
a view encoder configured, using a processor, to:
encode in the data stream, first information associated with a first coding block in a first view of the plurality of views of the multi-view video signal, wherein the first information is indicative of whether a first motion parameter of the first coding block of the first view is to:
(a) be encoded in the data stream independent of another coding parameter, or
(b) reuse one or more coding parameters of a second coding block located in a second view of the plurality of views of the multi-view video signal;
if the first information indicates that the first motion parameter of the first coding block of the first view is to be read from the data stream independent of another coding parameter, encode in the data stream, coding parameters including the first motion parameter of the first block of the first view;
if the first information indicates that the first motion parameter of the first coding block of the first view is to reuse one or more coding parameters of the second coding block located in the second view, encode in the data stream, the one or more coding parameters including a second motion parameter of the second coding block, wherein the second motion parameter is to be adopted as the first motion parameter; and
encode in the data stream, residual data associated with the first coding block of the first view of the multi-view video signal.
11 . The encoder of claim 10 , wherein if the first information indicates that the first motion parameter of the first coding block is to reuse one or more coding parameters of the second coding block located in the second view, the view encoder is configured to encoder in the data stream, second information identifying the second view from among the plurality of views of the multi-view video signal.
12 . The encoder of claim 11 , wherein the second information includes one or more reference frame indexes of the second view.
13 . The encoder of claim 10 , further configured for using the processor to encode an indication in the data stream indicating whether the first coding block is encoded using one of intra-prediction or inter-prediction.
14 . The encoder of claim 13 , wherein the indication of inter-prediction is inter-frame prediction.
15 . The encoder of claim 10 , wherein the prediction and residual of the first coding block is encoded using a different special resolution than the second coding block.
16 . The encoder of claim 10 , wherein the first coding block includes one of video data and a depth map.
17 . The encoder of claim 16 , wherein, if the first information indicates that the first motion parameter of the first coding block of the first view is to reuse one or more coding parameters of the second coding block, a prediction of the video data is encoded based on a first subset of the one or more coding parameters of the second coding block.
18 . The encoder of claim 16 , wherein, if the first information indicates that the first motion parameter of the first coding block of the first view is to reuse one or more coding parameters of the second coding block, a prediction of the depth map is encoded based on a second subset of the one or more coding parameters of the second coding block.
19 . A method for encoding a multi-view video into a data stream, the multi-view video including a plurality of views, comprising:
encoding, in the data stream, first information associated with a first coding block in a first view of the plurality of views of the multi-view video, wherein the first information is indicative of whether a first motion parameter of the first coding block of the first view is to:
(a) be encoded in the data stream independent of another coding parameter, or
(b) reuse one or more coding parameters of a second coding block located in a second view of the plurality of views of the multi-view video;
if the first information indicates that the first motion parameter of the first coding block of the first view is to be read from the data stream independent of another coding parameter, encoding, in the data stream, coding parameters including the first motion parameter of the first coding block of the first view;
if the first information indicates that the first motion parameter of the first coding block of the first view is to reuse one or more coding parameters of the second coding block located in the second view, encoding, in the data stream, the one or more coding parameters including a second motion parameter of the second coding block, wherein the second motion parameter is to be adopted as the first motion parameter; and
encoding, in the data stream, residual data associated with the first coding block of the first view of the multi-view video.
20 . A method for storing a data stream into which information related to a multi-view video is encoded according to the method of claim 19 .
21 . A method for transmitting a data stream into which information related to a multi-view video is encoded according to the method of claim 19 .