Multi-pass add-on tool for coherent and complete view synthesis

Patent metadata

Field	Value
Publication number	US-10965932-B2
Application number	US-201916456197-A
Country	US
Kind code	B2
Filing date	Jun 28, 2019
Priority date	Mar 19, 2019
Publication date	Mar 30, 2021
Grant date	Mar 30, 2021

Abstract

An embodiment of an image processor for immersive video includes technology to re-order patches from a plurality of views based on one or more of relative position and orientation related information for a desired synthesized view, select a set of views to be used in each view synthesis pass, perform two or more view synthesis passes for the synthesized view to provide two or more intermediate view synthesis results, and mask and merge the two or more intermediate view synthesis results to provide a final view synthesis result. Other embodiments are disclosed and claimed.

First claim

What is claimed is:

1. An electronic system, comprising:
memory to store one or more patches from a plurality of views;
a processor communicatively coupled to the memory; and
logic coupled to the processor, the logic to:
re-order the one or more patches from the plurality of views based on one or more of relative position and orientation related information for a desired synthesized view;
select a respective set of views to be used for each of two or more view synthesis passes;
perform the two or more view synthesis passes for the desired synthesized view using only patches belonging to the selected set of views for each pass to provide two or more intermediate view synthesis results; and
mask and merge the two or more intermediate view synthesis results to provide a final view synthesis result.

2. The system of claim 1 , wherein the logic is further to:
re-order the one or more patches from the plurality of views based on a distance between each view and the desired synthesized view.

3. The system of claim 2 , wherein the logic is further to:
determine overlapped scene coverage information between the plurality of views and the desired synthesized view based on the one or more of relative position and orientation related information; and
re-order the one or more patches from the plurality of views in an order from a relatively largest overlap to a relatively smallest overlap based on the determined overlapped scene coverage information between the plurality of views and the desired synthesized view.

4. The system of claim 1 , wherein the logic is further to:
perform two or more of the view synthesis passes in parallel with each other.

5. The system of claim 4 , wherein the logic is further to:
determine one or more of a number of passes and a number of views for the performed view synthesis based on one or more user-provided parameters.

6. The system of claim 1 , wherein the logic is further to:
generate a binary mask to identify missing information in an intermediate view.

7. The system of claim 6 , wherein the logic is further to:
merge two intermediate view synthesis results based on the binary mask such that missing information is taken from the higher pass intermediate synthesized view and coherent synthesis information is taken from the lower pass intermediate synthesized view; and
repeat the merge consecutively for each intermediate synthesized view result.

8. A method of processing an image, comprising:
re-ordering one or more patches from a plurality of views based on one or more of relative position and orientation related information for a desired synthesized view;
selecting a respective set of views to be used for each of two or more view synthesis passes;
performing the two or more view synthesis passes for the desired synthesized view using only patches belonging to the selected set of views for each pass to provide two or more intermediate view synthesis results; and
masking and merging the two or more intermediate view synthesis results to provide a final view synthesis result.

9. The method of claim 8 , further comprising:
re-ordering the one or more patches from the plurality of views based on a distance between each view and the desired synthesized view.

10. The method of claim 9 , further comprising:
determining overlapped scene coverage information between the plurality of views and the desired synthesized view based on the one or more of relative position and orientation related information; and
re-ordering the one or more patches from the plurality of views in an order from a relatively largest overlap to a relatively smallest overlap based on the determined overlapped scene coverage information between the plurality of views and the desired synthesized view.

11. The method of claim 8 , further comprising:
performing two or more of the view synthesis passes in parallel with each other.

12. The method of claim 11 , further comprising:
determining one or more of a number of passes and a number of views for the performed view synthesis based on one or more user-provided parameters.

13. The method of claim 8 , further comprising:
generating a binary mask to identify missing information in an intermediate view.

14. The method of claim 13 , further comprising:
merging two intermediate view synthesis results based on the binary mask such that missing information is taken from the higher pass intermediate synthesized view and coherent synthesis information is taken from the lower pass intermediate synthesized view; and
repeating the merge consecutively for each intermediate synthesized view result.

15. At least one non-transitory machine readable medium comprising a plurality of instructions that, in response to being executed on a computing device, cause the computing device to:
re-order one or more patches from a plurality of views based on one or more of relative position and orientation related information for a desired synthesized view;
select a respective set of views to be used for each of two or more view synthesis passes;
perform the two or more view synthesis passes for the desired synthesized view using only patches belonging to the selected set of views for each pass to provide two or more intermediate view synthesis results; and
mask and merge the two or more intermediate view synthesis results to provide a final view synthesis result.

16. The machine readable medium of claim 15 , comprising a plurality of further instructions that, in response to being executed on the computing device, cause the computing device to:
re-order the one or more patches from the plurality of views based on a distance between each view and the desired synthesized view.

17. The machine readable medium of claim 16 , comprising a plurality of further instructions that, in response to being executed on the computing device, cause the computing device to:
determine overlapped scene coverage information between the plurality of views and the desired synthesized view based on the one or more of relative position and orientation related information; and
re-order the one or more patches from the plurality of views in an order from a relatively largest overlap to a relatively smallest overlap based on the determined overlapped scene coverage information between the plurality of views and the desired synthesized view.

18. The machine readable medium of claim 15 , comprising a plurality of further instructions that, in response to being executed on the computing device, cause the computing device to:
perform two or more of the view synthesis passes in parallel with each other.

19. The machine readable medium of claim 18 , comprising a plurality of further instructions that, in response to being executed on the computing device, cause the computing device to:
determine one or more of a number of passes and a number of views for the performed view synthesis based on one or more user-provided parameters.

20. The machine readable medium of claim 15 , comprising a plurality of further instructions that, in response to being executed on the computing device, cause the computing device to:
generate a binary mask to identify missing information in an intermediate view.

21. The machine readable medium of claim 20 , comprising a plurality of further instructions that, in response to being executed on the computing device, cause the computing device to:
merge two intermediate view synthesis results based on the binary mask such that missing infor

Assignees

• Intel Corp

Inventors

• Salahieh Basel
• Bhatia Sumit
• Boyce Jill

Classifications

• H04N13/344

  with head-mounted left-right displays · CPC title

• H04N13/351

  for displaying simultaneously · CPC title

• H04N19/597

  specially adapted for multi-view video sequence encoding · CPC title

• H04N13/156

  Mixing image signals · CPC title

• H04N13/282 Primary

  for generating image signals corresponding to three or more geometrical viewpoints, e.g. multi-view systems · CPC title