Systems, methods, and media for generating visualization of physical environment in artificial reality
In one embodiment for generating passthrough, a system may receive an image and depth measurements of an environment and generate a corresponding 3D model. The system identifies, in the image, first pixels depicting a physical object and second pixels corresponding to a padded boundary around the first pixels. The system associates the first pixels with a first portion of the 3D model representing the physical object and a first representative depth value computed based on the depth measurements. The system associates the second pixels with a second portion of the 3D model representing a region around the physical object and a second representative depth value farther than the first representative depth value. The system renders an output image depicting a virtual object and the physical object. Occlusions between the virtual object and the physical object are determined using the first representative depth value and the second representative depth value.
1. A method comprising, by a computing system:
receiving an image and depth measurements of an environment surrounding a user;
generating a 3D model representing the environment based on the depth measurements;
identifying, in the image, first pixels depicting a physical object in the environment and second pixels corresponding to a padded boundary around the first pixels;
associating the first pixels with (1) a first portion of the 3D model representing the physical object and (2) a first representative depth value computed based on the depth measurements;
associating the second pixels with (1) a second portion of the 3D model representing a region around the physical object and (2) a second representative depth value farther than the first representative depth value; and
rendering an output image depicting a virtual object and the physical object, wherein occlusions between the virtual object and the physical object are determined using the first representative depth value and the second representative depth value.
2. The method of claim 1 , wherein the second representative depth value is computed based on the first representative depth value.
3. The method of claim 1 , wherein the first pixels are identified by processing the image using a machine-learning model.
4. The method of claim 1 , wherein the second pixels are identified based on a gradient of the first pixels.
5. The method of claim 1 , wherein:
the output image depicts a first portion of the virtual object being occluded by the physical object, and
an occlusion of the first portion of the virtual object by the physical object is determined based on a comparison between depth values associated with the first portion of the virtual object and the first representative depth value.
6. The method of claim 5 , wherein:
the output image depicts a second portion of the virtual object occluding a portion of the region around the physical object, and
an occlusion of the portion of the region around the physical object by the second portion of the virtual object is determined based on a comparison between depth values associated with the second portion of the virtual object and the second representative depth value.
7. The method of claim 1 , wherein the occlusions between the virtual object and the physical object are determined by:
determining, for a pixel of the output image, that a gaze from a viewpoint intersects the first portion of the 3D model representing the physical object and a portion of the virtual object; and
determining an occlusion between the first portion of the 3D model and the portion of the virtual object by comparing the first representative depth value associated with the first portion of the 3D model and a depth value associated with the portion of the virtual object.
8. The method of claim 1 , wherein the occlusions between the virtual object and the physical object are determined by:
determining, for a pixel of the output image, that a gaze from a viewpoint intersects the second portion of the 3D model representing the region around the physical object and a portion of the virtual object; and
determining an occlusion between the second portion of the 3D model and the portion of the virtual object by comparing the second representative depth value associated with the second portion of the 3D model and a depth value associated with the portion of the virtual object.
9. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
receive an image and depth measurements of an environment surrounding a user;
generate a 3D model representing the environment based on the depth measurements;
identify, in the image, first pixels depicting a physical object in the environment and second pixels corresponding to a padded boundary around the first pixels;
associate the first pixels with (1) a first portion of the 3D model representing the physical object and (2) a first representative depth value computed based on the depth measurements;
associate the second pixels with (1) a second portion of the 3D model representing a region around the physical object and (2) a second representative depth value farther than the first representative depth value; and
render an output image depicting a virtual object and the physical object, wherein occlusions between the virtual object and the physical object are determined using the first representative depth value and the second representative depth value.
10. The one or more computer-readable non-transitory storage media of claim 9 , wherein the second representative depth value is computed based on the first representative depth value.
11. The one or more computer-readable non-transitory storage media of claim 9 , wherein the first pixels are identified by processing the image using a machine-learning model.
12. The one or more computer-readable non-transitory storage media of claim 9 , wherein the second pixels are identified based on a gradient of the first pixels.
13. The one or more computer-readable non-transitory storage media of claim 9 , wherein:
the output image depicts a first portion of the virtual object being occluded by the physical object, and
an occlusion of the first portion of the virtual object by the physical object is determined based on a comparison between depth values associated with the first portion of the virtual object and the first representative depth value.
14. The one or more computer-readable non-transitory storage media of claim 13 , wherein:
the output image depicts a second portion of the virtual object occluding a portion of the region around the physical object, and
an occlusion of the portion of the region around the physical object by the second portion of the virtual object is determined based on a comparison between depth values associated with the second portion of the virtual object and the second representative depth value.
15. A system comprising:
one or more processors; and
one or more computer-readable non-transitory storage media coupled to one or more of the processors and comprising instructions operable when executed by one or more of the processors to cause the system to:
receive an image and depth measurements of an environment surrounding a user;
generate a 3D model representing the environment based on the depth measurements;
identify, in the image, first pixels depicting a physical object in the environment and second pixels corresponding to a padded boundary around the first pixels;
associate the first pixels with (1) a first portion of the 3D model representing the physical object and (2) a first representative depth value computed based on the depth measurements;
associate the second pixels with (1) a second portion of the 3D model representing a region around the physical object and (2) a second representative depth value farther than the first representative depth value; and
render an output image depicting a virtual object and the physical object, wherein occlusions between the virtual object and the physical object are determined using the first representative depth value and the second representative depth value.
16. The system of claim 15 , wherein the second representative depth value is computed based on the first representative depth value.
17. The system of claim 15 , wherein the first pixels are identified by processing the image using a machine-learning model.
18. The system of claim 15 , wherein the second pixels are identified based on a gradient of the first pixels.
19. The system of claim 15 , wherein:
the output image depicts a first portion of the virtual object being occluded by the physical object, and
an occlusion of the first portion of the virtual object by the physical object is determined based on a comparison between depth values associated with the first portion of the virtual object and the first representative depth value.
20. The system of claim 19 , wherein:
the output image depicts a second portion of the virtual object occluding a portion of the region around the physical object, and
an occlusion of the portion of the region around the physical object by the second portion of the virtual object is determined based on a comparison between depth values associated with the second portion of the virtual object and the second representative depth value.