Back to results
US 10,097,810 B2Granted
Efficient multi-view coding using depth-map estimate and update
AssigneeGe Video Compression, LLC
InventorsHeiko Schwarz, Thomas Wiegand
Life of this patent
Estimating the term…
Abstract
The unavailability of a depth map for a current picture of a reference view may be adequately addressed so as to reduce inter-view redundancies by estimating a depth map for the pictures of the reference and dependent views and updating same using motion and/or disparity data signaled within the multi-view data stream.
Claims30
A method for reconstructing a multi-view signal coded in a data stream, comprising:
- processing a list of plurality of motion vector candidates associated with a coding block of a current picture in a dependent view of the multi-view signal, comprising:
- obtaining a disparity vector with respect to the coding block of the current picture in the dependent view, wherein the disparity vector represents a disparity between the current picture of the dependent view and a current picture of a reference view of the multi-view signal,
- identifying a corresponding block in the current picture of the reference view based on the disparity vector,
- estimating a first motion vector based on a second motion vector associated with the corresponding block in the current picture of the reference view,
- adding the first motion vector into the list of plurality of motion vector candidates, and
- extracting, from the data stream, index information that specifies at least one of the list of motion vector candidates to be used for motion-compensated prediction;
- and
- reconstructing the coding block in the current picture of the dependent view by performing the motion-compensated prediction based on the at least one of the list of motion vector candidates.
5 more elements
An apparatus for reconstructing a multi-view signal coded in a data stream, comprising:
- a dependent view reconstructor configured for:
- processing a list of plurality of motion vector candidates associated with a coding block of a current picture in a dependent view of the multi-view signal, comprising:
- obtaining a disparity vector with respect to the coding block of the current picture in the dependent view, wherein the disparity vector represents a disparity between the current picture of the dependent view and a current picture of a reference view of the multi-view signal,
- identifying a corresponding block in the current picture of the reference view based on the disparity vector,
- estimating a first motion vector based on a second motion vector associated with the corresponding block in the current picture of the reference view,
- adding the first motion vector into the list of plurality of motion vector candidates, and
- extracting, from the data stream, index information that specifies at least one of the list of motion vector candidates to be used for motion-compensated prediction;
- and
- reconstructing the coding block in the current picture of the dependent view by performing the motion-compensated prediction based on the at least one of the list of motion vector candidates.
6 more elements
Machine readable non-transitory medium having data stored thereon for reconstructing a multi-view signal coded in a data stream, wherein the data, when read by a machine, cause the machine to at least perform:
- processing a list of plurality of motion vector candidates associated with a coding block of a current picture in a dependent view of the multi-view signal, comprising:
- obtaining a disparity vector with respect to the coding block of the current picture in the dependent view, wherein the disparity vector represents a disparity between the current picture of the dependent view and a current picture of a reference view of the multi-view signal,
- identifying a corresponding block in the current picture of the reference view based on the disparity vector,
- estimating a first motion vector based on a second motion vector associated with the corresponding block in the current picture of the reference view,
- adding the first motion vector into the list of plurality of motion vector candidates, and
- extracting, from the data stream, index information that specifies at least one of the list of motion vector candidates to be used for motion-compensated prediction;
- and
- reconstructing the coding block in the current picture of the dependent view by performing the motion-compensated prediction based on the at least one of the list of motion vector candidates.
5 more elements
A method for encoding a multi-view signal into a data stream, comprising:
- processing a list of plurality of motion vector candidates associated with a coding block of a current picture in a dependent view of the multi-view signal, comprising:
- obtaining a disparity vector with respect to the coding block of the current picture in the dependent view, wherein the disparity vector represents a disparity between the current picture of the dependent view and a current picture of a reference view of the multi-view signal,
- identifying a corresponding block in the current picture of the reference view based on the disparity vector,
- estimating a first motion vector based on a second motion vector associated with the corresponding block in the current picture of the reference view, and
- adding the first motion vector into the list of plurality of motion vector candidates for the coding block,
- determining at least one of the list of motion vector candidates to be used for reconstructing the coding block in the current picture of the dependent view via motion-compensated prediction, and
- generating index information specifying the at least one of the list of motion vector candidates;
- and
- inserting, into the data stream, the index information.
6 more elements
An apparatus for encoding a multi-view signal coded into a data stream, comprising:
- a dependent view encoder configured for:
- processing a list of plurality of motion vector candidates associated with a coding block of a current picture in a dependent view of the multi-view signal, comprising:
- obtaining a disparity vector with respect to the coding block of the current picture in the dependent view, wherein the disparity vector represents a disparity between the current picture of the dependent view and a current picture of a reference view of the multi-view signal,
- identifying a corresponding block in the current picture of the reference view based on the disparity vector,
- estimating a first motion vector based on a second motion vector associated with the corresponding block in the current picture of the reference view,
- adding the first motion vector into the list of plurality of motion vector candidates,
- determining at least one of the list of motion vector candidates to be used for reconstructing the coding block in the current picture of the dependent view via motion-compensated prediction, and
- generating index information specifying the at least one of the list of motion vector candidates;
- and
- inserting, into the data stream, the index information.
7 more elements
Machine readable non-transitory medium having data stored thereon for encoding a multi-view signal into a data stream, wherein the data, when read by a machine, cause the machine to perform the following:
- processing a list of plurality of motion vector candidates associated with a coding block of a current picture in a dependent view of the multi-view signal, comprising:
- obtaining a disparity vector with respect to the coding block of the current picture in the dependent view, wherein the disparity vector represents a disparity between the current picture of the dependent view and a current picture of a reference view of the multi-view signal,
- identifying a corresponding block in the current picture of the reference view based on the disparity vector,
- estimating a first motion vector based on a second motion vector associated with the corresponding block in the current picture of the reference view,
- adding the first motion vector into a list of plurality of motion vector candidates for the coding block,
- determining at least one of the list of motion vector candidates to be used for reconstructing the coding block in the current picture of the dependent view via motion-compensated prediction, and
- generating index information specifying the at least one of the list of motion vector candidates;
- and
- inserting, into the data stream, the index information.
6 more elements
Dependencies are inferred from the claim text.