Granted Patent B2

US 12,231,608 · App. 17/531,478 · Granted Feb 18, 2025

Inventors: Heiko Schwarz (Panketal, DE); Thomas Wiegand (Berlin, DE); Martin Winken (Berlin, DE)

Assignee: Dolby Video Compression, LLC

H04N13/128H04N13/161H04N19/463H04N19/597H04N2213/003

Abstract

Claims (24)

1. A method for reconstructing a multi-view signal coded in a data stream, comprising:

obtaining a disparity vector representing a view displacement between a current view and a reference view of a multi-view signal;

identifying a reference-view coding block in a picture of the reference view based on the disparity vector;

obtaining reference motion information associated with the reference-view coding block;

determining current motion information associated with a current coding block in a picture of the current view based on the reference motion information; and

reconstructing the current coding block based on the current motion information and a current-view residual signal associated with the current coding block, wherein the current-view residual signal is based on a reference-view residual signal associated with the reference-view coding block.

2. The method of claim 1 , further comprising obtaining the current-view residual signal based on the disparity vector.

3. The method of claim 2 , wherein obtaining the current-view residual signal is based on the disparity vector and the reference-view residual signal associated with the reference-view coding block.

4. The method of claim 3 , wherein the reference-view residual signal and the current-view residual signal are represented in a transform domain.

5. The method of claim 1 , further comprising:

extracting, from the data stream, a disparity vector residual signal;

refining the disparity vector based on the disparity vector residual signal to generate a refined disparity vector; and

identifying the reference-view coding block based on the refined disparity vector.

6. The method of claim 1 , wherein the obtaining of the disparity vector includes estimating the disparity vector based on temporal prediction using information related to a block in a previously reconstructed picture of the current view.

7. A method for encoding a multi-view signal into a data stream, comprising:

obtaining a disparity vector representing a view displacement between a current view and a reference view of a multi-view signal;

identifying a reference-view coding block in a picture of the reference view based on the disparity vector;

determining reference motion information associated with the reference-view coding block based on block-based predictive coding;

determining current motion information associated with a current coding block in a picture of the current view based on the reference motion information; and

inserting into the data stream the disparity vector and the reference motion information, wherein a current-view residual signal associated with the current coding block is based on a reference-view residual signal associated with the reference-view coding block.

8. The method of claim 7 , further comprising determining a reference-view residual signal associated with the reference-view coding block based on the block-based predictive coding; and inserting into the data stream the reference-view residual signal.

9. The method of claim 7 , wherein the current-view residual signal is determined based on the reference-view residual signal and the disparity vector.

10. The method of claim 7 , wherein the reference-view residual signal and the current-view residual signal are represented in a transform domain.

11. The method of claim 7 , wherein the obtaining of the disparity vector includes estimating the disparity vector based on temporal prediction using information related to a block in a previously reconstructed picture of the current view.

Assignments (2)

Continuity (5)

References Cited (168)