Method of determining an image distribution for a light field data structure
View Patent ↗The invention relates to a method of determining an image distribution (D opt ) for a light field data structure, which method comprises obtaining a plurality of images (F 1 , F 2 , . . . , F n ) from a plurality of image sources (C 1 , C 2 , . . . , C n ), performing image analysis on each image (F 1 , F 2 , . . . , F n ) of the plurality of images (F 1 , F 2 , . . . , F n ) to determine whether a specified criterion is satisfied by the content of that image (F 1 , F 2 , . . . , F n ), and identifying a group ( 12 ) of images (F 1 , F 2 , . . . , F n ) whose contents satisfy the specified criterion. The image group ( 12 ) is compared to each reference image distribution (D 1 , D 2 , . . . , D m ) of a set of predefined reference image distributions (D 1 , D 2 , . . . , D m ,) to select an optimal image distribution (D opt ), wherein a reference image distribution (D 1 , D 2 , . . . , D m ) comprises a predefined arrangement of I-images and P-images of the light field data structure. Each image (F 1 , F 2 , . . . , F n ) of the plurality of images (F 1 , F 2 , . . . , F n ) of the light field data structure is subsequently designated to be either an I-image or a P-image according to the selected image distribution (D opt ). The invention also describes a system ( 1 ) for determining an image distribution (D opt ) for a light field data structure.
1. A method of determining an image distribution for a light field data structure, which method comprises
obtaining a plurality of images from a plurality of image sources;
performing image analysis on each image of the plurality of images to determine whether a specified criterion is satisfied by content of that image;
identifying a group of images whose contents satisfy the specified criterion;
comparing the group of images to each reference image distribution of a set of predefined reference image distributions to select an optimal image distribution, wherein a reference image distribution includes a predefined arrangement of I-images and P-images of the light field data structure; and
designating each image of the plurality of images of the light field data structure to be either an I-image or a P-image according to the selected image distribution.
2. The method according to claim 1 , wherein the specified criterion includes parameters related to an object to be detected in content of an image.
3. The method according to claim 1 , wherein the step of identifying a group of images includes compiling an image-related array based on the images that satisfy the specified criterion.
4. The method according to claim 1 , wherein the optimal image distribution corresponds to the reference image distribution which provides a greatest number of I-images for the image group.
5. The method according to claim 1 , wherein the step of performing image analysis includes running a segmentation algorithm on that image.
6. A method of performing video coding of a light field data structure, which method comprises
determining an optimal image distribution for images of the light field data structure using the method of claim 1 ;
coding each image designated to be an I-image using a compression algorithm;
and coding each image designated to be a P-image using a previously coded image.
7. A system for determining an image distribution for a light field data structure, comprising
a plurality of image sources for obtaining a plurality of images;
an image analysis module for performing image analysis on each of the plurality of the plural images to determine whether a specified criterion is satisfied by content of that image;
an image group identification module for identifying a group of images whose contents satisfy the specified criterion;
a comparison module for comparing the image group to each reference image distribution of a set of predefined reference image distributions to select an optimal image distribution, wherein a reference image distribution includes a predefined arrangement of I-images and P-images of the light field data structure; and
an image designation module for designating each of the plural images of the light field data structure to be either an I-image or a P-image based on the selected image distribution.
8. A system for determining an image distribution for a light field data structure according to claim 7 , further comprising an interface for specifying a criterion.
9. A non-transitory computer readable medium encoded with data and instructions that when executed by a computer causes a computer to:
obtain a plurality of images from a plurality of image sources;
perform image analysis on each image of the plurality of images to determine whether a specified criterion is satisfied by content of that image;
identify a group of images whose contents satisfy the specified criterion;
compare the group of images to each reference image distribution of a set of predefined reference image distributions to select an optimal image distribution, wherein a reference image distribution includes a predefined arrangement of I-images and P-images of the light field data structure; and
designate each image of the plurality of images of the light field data structure to be either an I-image or a P-image according to the selected image distribution.
10. A system for performing video coding, comprising:
a system for determining an image distribution for a light field data structure comprising:
a plurality of image sources for obtaining a plurality of images;
an image analysis module for performing image analysis on each of the plurality of the plural images to determine whether a specified criterion is satisfied by content of that image;
an image group identification module for identifying a group of images whose contents satisfy the specified criterion;
a comparison module for comparing the image group to each reference image distribution of a set of predefined reference image distributions to select an optimal image distribution, wherein a reference image distribution includes a predefined arrangement of I-images and P-images of the light field data structure; and
an image designation module for designating each of the plural images of the light field data structure to be either an I-image or a P-image based on the selected image distribution;
an I-image coder for coding an image, designated to be an I-image, using a compression algorithm; and
a P-image coder for coding an image, designated to be a P-image, using a previously coded image.