Method and apparatus for rendering facades of objects of interest from three-dimensional point clouds

1. A method comprising:

obtaining a plurality of three-dimensional (3D) point clouds about a plurality of objects of interest, each of said 3D point clouds being labelled to a category of objects of interest;

rendering facades for the objects of interest categorized as buildings using an ambient occlusion method, where illumination of the point cloud is calculated based on light coming from a theoretical hemisphere or sphere around the object of interest; and

rendering shapes of the objects of interest categorized as non-buildings by fitting predefined templates of street view objects to the point clouds labelled as non-buildings.

2. The method according to claim 1 , wherein the predefined templates of street view objects are retrieved from a library of meshes providing a variety of mesh structures descriptive of various non-building street view objects.

3. The method according to claim 1 , further comprising

dividing categories of the street view objects into two subsets; and

adopting different template fitting approaches to a first and a second subset of street view object categories.

4. The method according to claim 3 , wherein the first subset includes street view objects, for which the orientation of the mesh structure is irrelevant such that their object models are definable based on a position and dimensions of the object.

5. The method according to claim 4 , further comprising

calculating, for the separated point cloud, the center of the point cloud and its boundaries;

selecting, based on the size of the meshes in the library, an isodiametric mesh for said point cloud, and

fitting the selected mesh by matching its center to the center of the point cloud and stretching the mesh to an appropriate size.

6. The method according to claim 4 , wherein the first subset of street view objects includes at least trees, persons, and sign symbols.

7. The method according to claim 3 , wherein the second subset includes street view objects, for which the orientation of the mesh structure, the position and the dimensions of the object are relevant for defining their object models.

8. The method according to claim 7 , further comprising

determining a bounding box around the object;

calculating, for the separated point cloud, the center of the point cloud and its boundaries; and

selecting, based on the dimension of the bounding box of the object, an isodiametric mesh for said point cloud.

9. The method according to claim 7 , wherein the second subset of the street view objects includes at least vehicles in general including cars, busses, or bikes.

10. An apparatus comprising at least one processor, and memory including computer program code, the memory and the computer program code configured to, with the at least one processor, cause the apparatus to at least:

obtain a plurality of three-dimensional (3D) point clouds about a plurality of objects of interest, each of said 3D point clouds being labelled to a category of objects of interest;

render facades for the objects of interest categorized as buildings using an ambient occlusion method, where illumination of the point cloud is calculated based on light coming from a theoretical hemisphere or sphere around the object of interest; and

render shapes of the objects of interest categorized as non-buildings by fitting predefined templates of street view objects to the point clouds labelled as non-buildings.

11. The apparatus according to claim 10 , further comprising computer program code configured to, with the at least one processor, cause the apparatus to

retrieve the predefined templates of street view objects from a library of meshes providing a variety of mesh structures descriptive of various non-building street view objects.

12. The apparatus according to claim 10 , further comprising computer program code configured to, with the at least one processor, cause the apparatus to

divide categories of the street view objects into two subsets; and

adopt different template fitting approaches to a first and a second subset of street view object categories.

13. The apparatus according to claim 12 , wherein the first subset includes street view objects, for which the orientation of the mesh structure is irrelevant such that their object models are definable based on a position and dimensions of the object.

14. The apparatus according to claim 13 , further comprising computer program code configured to, with the at least one processor, cause the apparatus to

calculate, for the separated point cloud, the center of the point cloud and its boundaries;

select, based on the size of the meshes in the library, an isodiametric mesh for said point cloud, and

fit the selected mesh by matching its center to the center of the point cloud and stretching the mesh to an appropriate size.

15. The apparatus according to claim 13 , wherein the first subset of street view objects includes at least trees, persons, and sign symbols.

16. The apparatus according to claim 12 , wherein the second subset includes street view objects, for which the orientation of the mesh structure, the position and the dimensions of the object are relevant for defining their object models.

17. The apparatus according to claim 16 , further comprising computer program code configured to, with the at least one processor, cause the apparatus to

determine a bounding box around the object;

calculate, for the separated point cloud, the center of the point cloud and its boundaries; and

select, based on the dimension of the bounding box of the object, an isodiametric mesh for said point cloud.

18. The apparatus according to claim 16 , wherein the second subset of the street view objects includes at least vehicles in general including cars, busses, or bikes.

19. A non-transitory computer readable storage medium stored with code thereon for use by an apparatus, which when executed by a processor, causes the apparatus to perform:

obtaining a plurality of three-dimensional (3D) point clouds about a plurality of objects of interest, each of said 3D point clouds being labelled to a category of objects of interest;

rendering facades for the objects of interest categorized as buildings using an ambient occlusion method, where illumination of the point cloud is calculated based on light coming from a theoretical hemisphere or sphere around the object of interest; and

rendering shapes of the objects of interest categorized as non-buildings by fitting predefined templates of street view objects to the point clouds labelled as non-buildings.

20. An apparatus comprising

means for obtaining a plurality of three-dimensional (3D) point clouds about a plurality of objects of interest, each of said 3D point clouds being labelled to a category of objects of interest;

means for rendering facades for the objects of interest categorized as buildings using an ambient occlusion method, where illumination of the point cloud is calculated based on light coming from a theoretical hemisphere or sphere around the object of interest; and

means for rendering shapes of the objects of interest categorized as non-buildings by fitting predefined templates of street view objects to the point clouds labelled as non-buildings.