Path planning for street level navigation in a three-dimensional environment, and applications thereof

1. A computer-implemented method for street-level navigation of a virtual camera in a three-dimensional environment, the virtual camera defining a perspective to display the three-dimensional environment, comprising:

(a) determining, by one or more computing devices, a target location in the three-dimensional environment, the three-dimensional environment including a plurality of points, each point corresponding to a location where a panoramic photograph was taken;

(b) determining, by one or more computing devices, a path specified by a discontinuous series of two or more of the plurality of points in the three-dimensional environment such that the path runs from a first position corresponding to an initial location of the virtual camera to a second position corresponding to the target location, and wherein at least a portion of the path is determined to run along a representation of a street;

(c) moving, by one or more computing devices, the virtual camera in the three-dimensional environment toward the target location along the path determined in (b) such that the virtual camera is moved to each point in the series of points in an order specified by the series, wherein the three-dimensional environment includes a plurality of building models, at least one building model in the plurality of building models is positioned between the initial location of the virtual camera and the target location, and the virtual camera avoids colliding with the at least one building model by moving along the path specified the series of points; and

(d) orienting, by one or more computing devices, the virtual camera to face the target location as the virtual camera moves along the path.

2. The method of claim 1 , wherein determining (b) comprises:

(i) determining the path to turn at an intersection in a representation of a street network in the three-dimensional environment, and

(ii) interpolating a spline at the turn in the path to smooth motion of the virtual camera at the turn.

3. The method of claim 1 , wherein determining (b) comprises determining a first point in the series of points to be a position of the panoramic photograph closest to the initial location of virtual camera that does not specify that the virtual camera move away from the target location.

4. The method of claim 1 , wherein determining (b) comprises determining a final point in the series of points to be a position of the panoramic photograph closest to the target location of virtual camera that does not specify that the virtual camera move past the target location.

5. The method of claim 1 , wherein determining (b) comprises determining the path not to exceed a maximum length value.

6. The method of claim 1 , wherein determining (a) comprises:

(i) receiving position data representing a position on a viewport selected by a user;

(ii) extending a ray based on the position selected by the user;

(iii) determining the target location to be an intersection of the ray with an object in the three-dimensional model.

7. The method of claim 1 , further comprising:

(e) before the virtual camera is moved in (c) such that a view frustum of the virtual camera captures three-dimensional model data along the path, prefetching the three-dimensional model data.

8. A system for street-level navigation of a virtual camera in a three-dimensional environment, the virtual camera defining a perspective to display the three-dimensional environment, comprising:

a target module, implemented on one or more computing devices, that determines a target location in the three-dimensional environment, the three-dimensional environment including a plurality of points, each point corresponding to a location where a panoramic photograph was taken;

a path planner module, implemented on one or more computing devices, that determines a path specified by a discontinuous series of two or more of the plurality of points in the three-dimensional environment such that the path runs from a first position corresponding to an initial location of the virtual camera to a second position corresponding to the target location, and wherein at least a portion of the path is determined to run along a representation of a street; and

a path motion module, implemented on one or more computing devices, that, after the path planner module determines the path, moves the virtual camera in the three-dimensional environment toward the target location along the path and orients the virtual camera to face the target location as the virtual camera moves along the path such that the virtual camera is moved to each point in the series of points in an order specified by the series, wherein the three-dimensional environment includes a plurality of building models, at least one building model in the plurality of building models is positioned between the initial location of the virtual camera and the target location, and the virtual camera avoids colliding with the at least one building model by moving along the path specified by the series of points.

9. The system of claim 8 , wherein the path planner module determines the path to turn at an intersection in a representation of a street network in the three-dimensional environment and interpolates a spline at the turn in the path to smooth motion of the virtual camera at the turn.

10. The system of claim 8 , wherein the path planner module determines a first point in the series of points to be a position of the panoramic photograph closest to the initial location of virtual camera that does not specify that the virtual camera move away from the target location.

11. The system of claim 8 , wherein the path planner module determines a final point in the series of points to be the position of the panoramic photograph closest to the target location of virtual camera that does not specify that the virtual camera move past the target location.

12. The system of claim 8 , wherein the path planner module determines the path not to exceed a maximum length value.

13. The system of claim 8 , wherein the target module receives position data representing a position on a viewport selected by a user, extends a ray based on the position selected by the user, and determines the target location to be an intersection of the ray with an object in the three-dimensional model.

14. The system of claim 8 , wherein the path motion module, before the virtual camera is moved such that a view frustum of the virtual camera captures three-dimensional model data along the path, prefetches the three-dimensional model data.

15. The system of claim 8 , further comprising:

at least one computing device that implements the target module, the path planner module, and the path motion module.

16. A computer-implemented method for street-level navigation of a virtual camera in a three-dimensional environment, the virtual camera defining a perspective to display the three-dimensional environment, comprising:

(a) determining, by one or more computing devices, a target location in the three-dimensional environment, the three-dimensional environment including a plurality of points, each point corresponding to a location where a panoramic photograph was taken;

(b) determining, by one or more computing devices, a path specified by a discontinuous series of two or more of the plurality of points in the three-dimensional environment such that the path runs from a first position corresponding to an initial location of the virtual camera to a second position corresponding to the target location, and wherein at least a portion of the path is determined to run along a representation of a real-world pathway;

(c) moving, by one or more computing devices, the virtual camera in the three-dimensional environment toward the target location along the path determined in (b) such that the virtual camera is moved to each point in the series of points in an order specified by the series, wherein the three-dimensional environment includes a plurality of building models, at least one building model in the plurality of building models is positioned between the initial location of the virtual camera and the target location, and the virtual camera avoids colliding with the at least one building model by moving along the path specified by the series of points; and

(d) orienting, by one or more computing devices, the virtual camera to face the target location as the virtual camera moves along the path.