Method and device for encoding and decoding a light field based image, and corresponding computer program product

Patent metadata

Field	Value
Publication number	US-10536718-B2
Application number	US-201615757417-A
Country	US
Kind code	B2
Filing date	Sep 5, 2016
Priority date	Sep 7, 2015
Publication date	Jan 14, 2020
Grant date	Jan 14, 2020

Abstract

A method for encoding a current focal stack is disclosed that comprises a set of images focused at a different focalization distance from one image to another. According to present disclosure, the method comprises: —encoding (31) information representing an image of the current focal stack, the image being selected in said current focal stack according to an image sharpness criterion, and reconstructing the image into a reconstructed image; —encoding (32) at least another image of the current focal stack by prediction from at least the reconstructed image.

First claim

The invention claimed is:

1. A method comprising:
encoding a current focal stack including a set of images of a same scene focused at different focalization distances from one image to another, wherein said encoding comprises:
encoding information representing an image of the current focal stack, said image being selected in said current focal stack according to an image sharpness criterion, and reconstructing said image into a reconstructed image; and
encoding at least one other image of the current focal stack by prediction from at least said reconstructed image.

2. The method according to claim 1 , wherein, said image sharpness criterion corresponds to a number of pixels in focus in an image.

3. The method according to claim 2 , wherein, the selected image in the current focal stack is the one that has the highest number of pixels in focus.

4. The method according to claim 3 , wherein the image comprising the highest number of pixels in focus is an all-in-focus image.

5. The method according to claim 4 , wherein said method further comprises:
combining at least said encoded information representing said image of the current focal stack selected according to said image sharpness criterion and said at least one other encoded image of the current focal stack into a signal representing said current focal stack; and
inserting, a map to reconstruct said all in focus image of said current focal stack, in said signal representing said current focal stack.

6. The method according to claim 1 , wherein encoding at least one other image of the current focal stack by prediction from at least said reconstructed image comprises encoding, for at least a pixel of said other image, a value from at least a value of a corresponding pixel collocated in said reconstructed image and from a blurring function depending on a depth position associated with said other image to encode.

7. The method according to claim 1 , wherein said current focal stack is part of a group of focal stacks captured with different viewing angles, said focal stacks of said group of focal stacks being ordered according to an orientation order of their viewing angles, and wherein said method further comprises:
determining an inter-stack image encoding order from said orientation order and from a viewing angle, of said current focal stack, called reference viewing angle,
encoding at least one pixel of an image of at least another focal stack of said group of focal stacks processed according to said inter-stack image encoding order from the reconstructed image of said current focal stack, said image of said at least another focal stack, called a reference image, being the image selected according to an image sharpness criterion.

8. The method according to claim 7 , wherein said pixel of said reference image is encoded from said reconstructed image of the current focal stack, captured with said reference viewing angle, and from information representing a disparity between said reference image and said reconstructed image of the current focal stack.

9. The method according to claim 7 , wherein said value of each pixel of said reference image, captured with a viewing angle different from said reference value of viewing angle, is encoded from said reconstructed image of the current focal stack, captured with said reference value of viewing angle, and from a layered depth image of a scene associated to said group of focal stacks.

10. The method according to claim 7 , wherein said group of focal stacks is captured at a reference instant t 0 , and is a part of a sequence further comprising at least one other group of focal stacks captured at an instant t, said method further comprises, encoding at least one pixel of a reference image of a focal stack, captured with said reference value of viewing angle, of said at least one other group of focal stacks, captured at an instant t, of said sequence of group of focal stacks, by using a temporal prediction performed from the reconstructed image of said current focal stack captured at said reference instant t 0 .

11. A method comprising:
decoding at least a current focal stack including a set of images of a same scene focused at different focalization distances from one image to another, wherein said decoding comprises:
decoding information representing an image of the current focal stack, said image fulfilling an image sharpness criterion,
reconstructing at least another image of the current focal stack by prediction from at least said decoded image.

12. The method according to claim 11 , wherein:
said image sharpness criterion corresponds to a number of pixels in focus in an image;
the selected image in the current focal stack is the one that has the highest number of pixels in focus; and
the image comprising the highest number of pixels in focus is an all-in-focus image.

13. The method according to claim 11 , wherein decoding at least one other image of the current focal stack by prediction from at least said reconstructed image comprises decoding, for at least a pixel of said other image, a value from at least a value of a corresponding pixel collocated in said reconstructed image and from a blurring function depending on a depth position associated with said other image to decode.

14. The method according to claim 11 , wherein said current focal stack is part of a group of focal stacks captured with different viewing angles, said focal stacks of said group of focal stacks being ordered according to an orientation order of their viewing angles, and wherein said method further comprises:
determining an inter-stack image decoding order from said orientation order and from a viewing angle, of said current focal stack, called reference viewing angle,
decoding at least one pixel of an image of at least another focal stack of said group of focal stacks processed according to said inter-stack image decoding order from the reconstructed image of said current focal stack, said image of said at least another focal stack, called a reference image, being the image selected according to an image sharpness criterion; wherein
said pixel of said reference image is decoded from said reconstructed image of the current focal stack, captured with said reference viewing angle, and from information representing a disparity between said reference image and said reconstructed image of the current focal stack.

15. A device comprising:
a memory; and
at least one processor coupled to the memory, the at least one processor being configured to:
encode a current focal stack comprising a set of images of a same scene focused at different focalization distances from one image to another, wherein said at least one processor being configured to encode the current focal stack comprises being further configured to:
encode information representing an image of the current focal stack, said image being selected in said current focal stack according to an image sharpness criterion, and reconstruct said image into a reconstructed image;
encode at least another image of the current focal stack by prediction from at least said reconstructed image.

16. The device according to claim 15 , wherein:
said image sharpness criterion corresponds to a number of pixels in focus in an image;
the selected image in the current focal stack is the one that has the highest number of pixels in focus; and
the image comprising the highest number of pixels in focus is an all-in focus image.

17. The device according to claim 15 , wherein the at least one processor being configure to encode at least one other image of the cu

Assignees

• Thomson Licensing
• Interdigital Vc Holdings Inc

Inventors

• Guillotel Philippe
• Thoreau Dominique
• Vandame Benoit
• Lopez Patrick
• Boisson Guillaume

Classifications

• H04N19/182

  the unit being a pixel · CPC title

• H04N19/162

  User input · CPC title

• H04N19/172

  the region being a picture, frame or field · CPC title

• H04N13/161

  Encoding, multiplexing or demultiplexing different image signal components (for multi-view video sequence encoding H04N19/597) · CPC title

• H04N19/105

  Selection of the reference unit for prediction within a chosen coding or prediction mode, e.g. adaptive choice of position and number of pixels used for prediction · CPC title