Space and content matching for augmented and mixed reality
Techniques for automatically placing and manipulating virtual objects of augmented reality (AR) and mixed reality (MR) simulations are described. One technique includes obtaining an indication of at least one virtual object available for placing within a real-world environment during an AR simulation or a MR simulation. A first representation of the real-world environment is generated, based on a scan of the real-world environment. At least one second representation of the real-world environment is generated from the first representation. A match is determined between the at least one virtual object and at least one available space within the real-world environment, based at least in part on evaluating the at least one virtual object and the at least one second representation with a machine learning model(s). The at least one virtual object is rendered on a computing device, based on the match.
1 . A computer-implemented method comprising:
obtaining an indication of at least one virtual object available for placing within a real-world environment as part of an augmented reality (AR) simulation or a mixed reality (MR) simulation;
generating a first representation of the real-world environment, based on a scan of the real-world environment with one or more sensors of a computing device, wherein the first representation includes a first three-dimensional (3D) mesh representing a geometry of the real-world environment;
generating at least one second representation of the real-world environment by computing an inverse of the first 3D mesh corresponding to the first representation of the real-world environment, wherein the at least one second representation comprises a second 3D mesh of at least one empty space in the real-world environment;
determining a match between one or more attributes of the at least one virtual object and one or more additional attributes of the at least one empty space within the real-world environment, based at least in part on evaluating the at least one virtual object and the at least one second representation of the real-world environment with one or more machine learning models; and
rendering the at least one virtual object at a position on a display of the computing device, based on the match, the position on the display being associated with the at least one empty space within the real-world environment.
2 . The computer-implemented method of claim 1 , wherein determining the match between the one or more attributes of the at least one virtual object and the one or more additional attributes of the at least one empty space within the real-world environment comprises determining a size of the at least one virtual object that will fit within the at least one empty space.
3 . The computer-implemented method of claim 2 , wherein rendering the at least one virtual object comprises rendering the at least one virtual object on the display of the computing device with the determined size.
4 . The computer-implemented method of claim 1 , wherein the at least one virtual object is a stationary virtual object.
5 . The computer-implemented method of claim 1 , wherein the at least one virtual object is a dynamic virtual object.
6 . The computer-implemented method of claim 1 , wherein generating the first representation of the real-world environment comprises filling in the first mesh within a y-up dimension associated with the real-world environment.
7 . A computer-implemented method comprising:
obtaining an indication of at least one virtual object available for placing within a real-world environment as part of an augmented reality (AR) simulation or a mixed reality (MR) simulation;
generating a first representation of the real-world environment, based on a scan of the real-world environment with one or more sensors of a computing device, wherein the first representation includes a first three-dimensional (3D) mesh representing a geometry of the real-world environment;
generating at least one second representation of the real-world environment by computing an inverse of the first 3D mesh corresponding to the first representation of the real-world environment, wherein the at least one second representation comprises a second 3D mesh of at least one empty space in the real-world environment;
determining a match between the at least one virtual object and the at least one empty space within the real-world environment, based at least in part on evaluating the at least one virtual object and the at least one second representation of the real-world environment with one or more machine learning models;
rendering the at least one virtual object at a position on a display of the computing device, based on the match, the position on the display being associated with the at least one empty space within the real-world environment; and
determining a plurality of paths of the computing device through the real-world environment,
wherein:
the at least one second representation indicates the at least one empty space along a first path of the plurality of paths, or
the at least one virtual object is matched to the at least one empty space along the first path of the plurality of paths.
8 . The computer-implemented method of claim 7 , wherein the plurality of paths comprises a first set of paths through the real-world environment based on a first height range and a second set of paths through the real-world environment based on a second height range.
9 . The computer-implemented method of claim 7 , wherein:
the at least one virtual object is matched to the at least one empty space along the first path of the plurality of paths;
rendering the at least one virtual object comprises receiving sensor data indicating a position of the computing device along the first path; and
the position on the display corresponds to the position of the computing device within the at least one empty space along the first path.
10 . A computing device comprising:
a display;
one or more sensors;
one or more processors; and
a memory storing instructions, which, when executed on the one or more processors perform an operation comprising:
obtaining an indication of at least one virtual object available for placing within a real-world environment as part of an augmented reality (AR) simulation or a mixed reality (MR) simulation;
generating a first representation of the real-world environment, based on a scan of the real-world environment with one or more sensors of a computing device, wherein the first representation includes a first three-dimensional (3D) mesh representing a geometry of the real-world environment;
generating at least one second representation of the real-world environment by computing an inverse of the first 3D mesh corresponding to the first representation of the real-world environment, wherein the at least one second representation comprises a second 3D mesh of at least one empty space in the real-world environment;
determining a match between one or more attributes of the at least one virtual object and one or more additional attributes of the at least one empty space within the real-world environment, based at least in part on evaluating the at least one virtual object and the at least one second representation of the real-world environment with one or more machine learning models; and
rendering the at least one virtual object at a position on the display, based on the match, the position on the display being associated with the at least one empty space within the real-world environment.
11 . The computing device of claim 10 , wherein determining the match between the one or more attributes of the at least one virtual object and the one or more additional attributes of the at least one empty space within the real-world environment comprises determining an orientation of the at least one virtual object that will fit within the at least one empty space.
12 . The computing device of claim 10 , the operation further comprising determining a plurality of paths of the computing device through the real-world environment, wherein the at least one second representation further indicates the at least one empty space along a first path of the plurality of paths.
13 . The computing device of claim 12 , wherein the plurality of paths comprises a first set of paths through the real-world environment based on a first height range and a second set of paths through the real-world environment based on a second height range.
14 . The computing device of claim 10 , further comprising determining a plurality of paths of the computing device through the real-world environment, wherein the one or more attributes of the at least one virtual object is matched to the one or more additional attributes of the at least one empty space along a first path of the plurality of paths.
15 . The computing device of claim 14 , wherein:
rendering the at least one virtual object comprises receiving sensor data indicating a position of the computing device along the first path; and
the position on the display corresponds to the position of the computing device within the at least one empty space along the first path.
16 . A non-transitory computer-readable medium containing computer program code that, when executed by operation of one or more computer processors, performs an operation comprising:
obtaining an indication of at least one virtual object available for placing within a real-world environment as part of an augmented reality (AR) simulation or a mixed reality (MR) simulation;
generating a first representation of the real-world environment, based on a scan of the real-world environment with one or more sensors of a computing device, wherein the first representation includes a first three-dimensional (3D) mesh representing a geometry of the real-world environment;
generating at least one second representation of the real-world environment by computing an inverse of the first 3D mesh corresponding to the first representation of the real-world environment, wherein the at least one second representation comprises a second 3D mesh of at least one empty space in the real-world environment;
determining a match between one or more attributes of the at least one virtual object and one or more additional attributes of the at least one empty space within the real-world environment, based at least in part on evaluating the at least one virtual object and the at least one second representation of the real-world environment with one or more machine learning models; and
rendering the at least one virtual object at a position on a display of the computing device, based on the match, the position on the display being associated with the at least one empty space within the real-world environment.
17 . The non-transitory computer-readable medium of claim 16 , wherein determining the match between the one or more attributes of the at least one virtual object and the one or more additional attributes of the at least one empty space within the real-world environment comprises:
determining a first class of the at least one virtual object that will fit within the at least one empty space;
performing semantic segmentation to identify a second class associated with the at least one empty space; and
matching the at least one virtual object to the at least one empty space based on the first class and the second class.
18 . The non-transitory computer-readable medium of claim 17 , wherein rendering the at least one virtual object comprises adjusting at least one of a size or an orientation of the at least one virtual object based on the at least one empty space.
19 . The non-transitory computer-readable medium of claim 16 , the operation further comprising determining a plurality of paths of the computing device through the real-world environment, wherein the at least one virtual object is matched to the at least one empty space along a first path of the plurality of paths.
20 . A computer-implemented method comprising:
obtaining an indication of at least one virtual object available for placing within a real-world environment as part of an augmented reality (AR) simulation or a mixed reality (MR) simulation;
generating a first representation of the real-world environment, based on a scan of the real-world environment with one or more sensors of a computing device, wherein the first representation includes a first three-dimensional (3D) mesh representing a geometry of the real-world environment;
determining a plurality of paths of the computing device through the real-world environment, wherein each path included in the plurality of paths includes a 3D volume of an amount of space taken by a user through the real-world environment;
generating at least one second representation of the real-world environment by computing an inverse of the first 3D mesh corresponding to the first representation of the real-world environment and the plurality of paths, wherein the at least one second representation indicates at least one available space along a first path of the plurality of paths;
determining a match between the at least one virtual object and the at least one available space within the real-world environment, based at least in part on evaluating the at least one virtual object and the at least one second representation of the real-world environment with one or more machine learning models; and
rendering the at least one virtual object at a position on a display of the computing device, based on the match, the position on the display being associated with the at least one available space within the real-world environment.
21 . The computer-implemented method of claim 20 , wherein the plurality of paths comprises a first set of paths through the real-world environment based on a first height range and a second set of paths through the real-world environment based on a second height range.
22 . The computer-implemented method of claim 20 , wherein the at least one virtual object is matched to the at least one available space along a second path of the plurality of paths.