US 12,347,022 B2
Method and system for generating a target image from plural multi-plane images
Andrew James Bigos, London (GB)
Assigned to Sony Interactive Entertainment Europe Limited, London (GB)
Appl. No. 18/017,793
Filed by Sony Interactive Entertainment Europe Limited, London (GB)
PCT Filed Jul. 23, 2021, PCT No. PCT/GB2021/051906
§ 371(c)(1), (2) Date Jan. 24, 2023,
PCT Pub. No. WO2022/018454, PCT Pub. Date Jan. 27, 2022.
Claims priority of application No. 2011546 (GB), filed on Jul. 24, 2020.
Prior Publication US 2023/0281912 A1, Sep. 7, 2023
Int. Cl. G06T 15/20 (2011.01); G06T 15/06 (2011.01); G06T 15/30 (2011.01); G06T 15/50 (2011.01)
CPC G06T 15/20 (2013.01) [G06T 15/06 (2013.01); G06T 15/30 (2013.01); G06T 15/503 (2013.01)] 25 Claims
OG exemplary drawing
 
1. A method of generating an image of a scene from a target viewpoint, the method comprising:
obtaining a three-dimensional (3D) representation of a virtual scene comprising at least one object;
capturing a plurality of multi-plane images of the 3D representation of the virtual scene, each multi-plane image corresponding to a different respective viewpoint of the 3D representation of the virtual scene;
wherein each multi-plane image comprises a plurality of image planes, each image plane comprising image data for a different respective successive depth segment of the virtual scene;
wherein each depth segment is defined by a pair of corresponding clipping planes that sub-divides a frustum of a virtual camera that captured-captures the corresponding multi-plane image;
obtaining virtual camera information associated with each of the captured multi-plane images;
determining, for each multi-plane image, whether at least one clipping plane in a respective clipping plane pair defined for that multi-plane image intersects an interior of an object within the 3D representation of the virtual scene;
responsive to a positive determination for at least one of the multi-plane images,
identifying image data in the corresponding image plane that corresponds with the intersected interior of the object and assigning an identifier to the image data to thereby obtain at least one respective modified multi-plane image;
obtaining camera information for a target virtual camera; and
generating an output image of the scene from a viewpoint of the target virtual camera by combining at least some of the multi-plane images together, at least one of the multi-plane images comprising a modified multi-plane image, said combination being performed in dependence on the obtained virtual camera information and target virtual camera information.