Occluder generation for structures in digital applications

Patent metadata

Field	Value
Publication number	US-12079925-B2
Application number	US-202318126047-A
Country	US
Kind code	B2
Filing date	Mar 24, 2023
Priority date	Jan 25, 2022
Publication date	Sep 3, 2024
Grant date	Sep 3, 2024

Abstract

A method is performed at a computing system for automatically generating an occluder, the method includes receiving an input model of the visual three-dimensional structure, the input model having a plurality of faces. The method includes simplifying the input model into an initial occluder including a plurality of candidate patches in a patch-based coarse mesh. The method includes determining a first quality metric of the initial occluder measured by a first number of pixels corresponding to objects behind the visual three-dimensional structure that are blocked by the input model and the initial occluder along a first view direction. The method includes removing one or more candidate patches associated with the first number of pixels from the initial occluder while maintaining the first quality metric above a first threshold to form the occluder for the visual three-dimensional structure.

First claim

What is claimed is:

1. A method for automatically generating an occluder for a visual three-dimensional structure, the method comprising:
receiving an input model of the visual three-dimensional structure, the input model comprising a plurality of faces;
simplifying the input model into an initial occluder including a plurality of candidate patches in a patch-based coarse mesh;
determining a first quality metric of the initial occluder associated with a precision, wherein the precision is measured by a first number of pixels corresponding to objects behind the visual three-dimensional structure that are blocked by the input model and the initial occluder along a first view direction; and
removing one or more candidate patches associated with the first number of pixels from the initial occluder while maintaining the first quality metric above a first threshold to form the occluder for the visual three-dimensional structure, wherein the first threshold is determined based on a change in a precision when a face from the initial occluder is removed, and the occluder blocks the objects behind the visual three-dimensional structure from being rendered in an application along the first view direction.

2. The method of claim 1 , wherein simplifying the input model into an initial occluder including a plurality of candidate patches in a patch-based coarse mesh comprises merging a first number of faces in the plurality of faces into the plurality of candidate patches, the plurality of candidate patches satisfying a threshold for the first quality metric.

3. The method of claim 2 , wherein merging the first number of faces into the plurality of candidate patches comprises combining faces within a first degree of coplanarity into a candidate planar patch in the plurality of candidate patches.

4. The method of claim 1 , further comprising:
voxelizing the input model to obtain a plurality of voxels for generating a voxelization-based coarse mesh; and
combining the patch-based coarse mesh and the voxelization-based coarse mesh to form the initial occluder.

5. The method of claim 4 , wherein the generating a voxelization-based coarse mesh includes:
voxelizing a bounding box of the input model to obtain the plurality of voxels;
computing a winding number for each of the plurality of voxels;
extracting an isosurface based on the winding number; and
simplifying the isosurface to obtain the voxelization-based coarse mesh.

6. The method of claim 1 , wherein the removing one or more candidate patches from the initial occluder comprises:
discretizing space around and within the input model into a plurality of equally sized blocks, each of the equally sized blocks having a plurality of view directions;
merging a portion of the plurality of equally sized blocks into a plurality of larger blocks; and
removing a view direction from a respective one of the plurality of larger blocks for which the input model is outside a view frustrum of the view direction.

7. The method of claim 1 , further comprising hole-filling the patch-based coarse mesh.

8. The method of claim 1 , wherein a number of faces in the occluder is less than 10% of a number of faces in the input model.

9. The method of claim 1 , wherein the first quality metric of the initial occluder is determined by identifying one or more pixels that are within a first two-dimensional area occluded by the input model of the visual three-dimensional structure and a second two-dimensional area occluded by the initial occluder along the first view direction.

10. A computing system comprising:
one or more processors;
memory; and
a plurality of programs stored in the memory, wherein the plurality of programs, when executed by the one or more processors, cause the computing system to perform a method for automatically generating an occluder for a visual three-dimensional structure, the method including:
receiving an input model of the visual three-dimensional structure, the input model comprising a plurality of faces;
simplifying the input model into an initial occluder including a plurality of candidate patches in a patch-based coarse mesh;
determining a first quality metric of the initial occluder associated with a precision, wherein the precision is measured by a first number of pixels corresponding to objects behind the visual three-dimensional structure that are blocked by the input model and the initial occluder along a first view direction; and
removing one or more candidate patches associated with the first number of pixels from the initial occluder while maintaining the first quality metric above a first threshold to form the occluder for the visual three-dimensional structure, wherein the first threshold is determined based on a change in a precision when a face from the initial occluder is removed, and the occluder blocks the objects behind the visual three-dimensional structure from being rendered in an application along the first view direction.

11. The computing system of claim 10 , wherein simplifying the input model into an initial occluder including a plurality of candidate patches in a patch-based coarse mesh comprises merging a first number of faces in the plurality of faces into the plurality of candidate patches, the plurality of candidate patches satisfying a threshold for the first quality metric.

12. The computing system of claim 11 , wherein merging the first number of faces into the plurality of candidate patches comprises combining faces within a first degree of coplanarity into a candidate planar patch in the plurality of candidate patches.

13. The computing system of claim 10 , wherein the method further comprises:
voxelizing the input model to obtain a plurality of voxels for generating a voxelization-based coarse mesh; and
combining the patch-based coarse mesh and the voxelization-based coarse mesh to form the initial occluder.

14. The computing system of claim 13 , wherein the generating a voxelization-based coarse mesh includes:
voxelizing a bounding box of the input model to obtain the plurality of voxels;
computing a winding number for each of the plurality of voxels;
extracting an isosurface based on the winding number; and
simplifying the isosurface to obtain the voxelization-based coarse mesh.

15. The computing system of claim 10 , wherein the removing one or more candidate patches from the initial occluder comprises:
discretizing space around and within the input model into a plurality of equally sized blocks, each of the equally sized blocks having a plurality of view directions;
merging a portion of the plurality of equally sized blocks into a plurality of larger blocks; and
removing a view direction from a respective one of the plurality of larger blocks for which the input model is outside a view frustrum of the view direction.

16. The computing system of claim 10 , wherein the method further comprises hole-filling the patch-based coarse mesh.

17. The computing system of claim 10 , wherein a number of faces in the occluder is less than 10% of a number of faces in the input model.

18. The computing system of claim 10 , wherein the first quality metric of the initial occluder is determined by identifying one or more pixels that are within a first two-dimensional area occluded by the input model of the visual three-dimensional structure and a second two-dimensional area occluded by the initial occluder along the first view direction.

19. A non-transitory computer readable storage medium storing a plurality of programs, wherein the plurality of

Assignees

• Tencent America LLC

Inventors

• Wu Kui
• Gao Xifeng
• Pan Zherong

Classifications

• G06T17/205 Primary

  Re-meshing · CPC title

• G06T15/40 Primary

  Hidden part removal · CPC title