Structure discovery in a point cloud

1. A structure discovery system, the system comprising:

a communication processor configured to receive 3-dimensional image data from an imaging device;

a surface discovery processor configured to rotate the 3-dimensional image data around axes in 3-dimensional space of the 3-dimensional image data to align points of at least one structure in the 3-dimensional image data and project the at least one structure onto a coordinate plane, wherein rotation of the 3-dimensional image data includes: i) rotating the 3-dimensional data through a first angle around a z axis, wherein rotation of the 3-dimensional data through a first angle orients facades of structures into a line projection, ii) rotating the 3-dimensional data through a second angle around a y-axis, wherein rotation of the 3-dimensional data around the second angle orients the facades of the structures into a vertical projection and iii) rotating the 3 -dimensional data through a third angle around the z-axis, wherein rotation of the 3-dimensional data through the third angle a) enables discovery of all facades of the structures, orientation of the facades, and b) enables separation and identification of foreground obscuring objects;

a linear condensation identification processor configured to identify linear condensations in the rotated 3-dimensional image data; and

a structure generation processor configured to generate the at least one structure based on the linear condensations and the rotated 3-dimensional image data.

2. The system of claim 1 , further comprising:

a structure identification processor configured to identify common structural elements in one or more surfaces from the 3-dimensional image data;

a point cloud rotation processor configured to rotate the one or more surfaces of the 3-dimensional image data based on the identified common structural elements; and

an image data merge processor configured to merge the one or more surfaces of the 3-dimensional image data to form one or more merged surfaces.

3. The system of claim 2 , wherein the one or more surfaces comprises a plane, a flat surface, a curved surface, or any combination thereof.

4. The system of claim 1 , further comprising:

a parallel line identification processor configured to identify one or more lines in the linear condensations; and

a structure identification processor configured to identify common structural elements in one or more surfaces from the 3-dimensional image data based on the one or more lines.

5. The system of claim 1 , wherein the 3-dimensional image data comprises a point cloud.

6. A method for structure discovery in a point cloud, the method comprising:

(a) receiving 3-dimensional image data;

(b) rotating the 3-dimensional image data around axes in 3-dimensional space of the 3-dimensional image data to align points of at least one structure in the 3-dimensional image data and project the at least one structure onto a coordinate plane, wherein rotating the 3-dimensional image data includes: i) rotating the 3-dimensional data through a first angle around a z-axis, wherein rotation of the 3-dimensional data through a first angle orients facades of structures into a line projection, ii) rotating the 3 -dimensional data through a second angle around a y-axis, wherein rotation of the 3-dimensional data around the second angle orients the facades of the structures into a vertical projections and iii) rotating the 3 -dimensional data through a third angle around the z-axis, wherein rotation of the 3-dimensional data through the third angle a) enables discovery of all facades of the structures, orientation of the facades, and b) enables separation and identification of foreground obscuring objects;

(c) identifying linear condensations in the rotated 3-dimensional image data; and

(d) generating the at least one structure based on the linear condensations and the rotated 3-dimensional image data.

7. The method of claim 6 , wherein orthogonal planes simultaneously project onto orthogonal viewing-planes during execution of step (b).

8. The method of claim 6 , wherein step (b) further comprises (b-1) projecting isometrically the 3-dimensional image data onto an observation plane.

9. The method of claim 6 , wherein step (b) further comprises (b-1) selecting the axis based on a plane of interest.

10. The method of claim 6 , wherein step (d) further comprises:

(d-1) determining a plurality of structural surfaces within the linear condensations;

(d-2) generating an image wireframe for each of the plurality of structural surfaces; and

(d-3) assembling the plurality of structural surfaces within the image wireframe to form the structure.

11. A non-transitory computer program product, tangibly embodied in a non-transitory information carrier, the non-transitory computer program product including instructions being operable to cause a data processing apparatus to:

receive 3-dimensional image data;

rotate the 3-dimensional image data around axes in 3-dimensional space of the 3-dimensional image data to align points of at least one structure in the 3-dimensional image data and project the at least one structure onto a coordinate plane, wherein rotation of the 3-dimensional image data includes: i) rotating the 3-dimensional data through a first angle around a z-axis, wherein rotation of the 3-dimensional data through a first angle orients facades of structures into a line projection, ii) rotating the 3 -dimensional data through a second angle around a y-axis, wherein rotation of the 3-dimensinoal data around the second angle orients the facades of the structures into a vertical projections and iii) rotating the 3 -dimensional data through a third angle around the z-axis, wherein rotation of the 3-dimensinoal data through the third angle a) enables discovery of all facades of the structures, orientation of the facades, and b) enables separation and identification of foreground obscuring objects;

identify linear condensations in the rotated 3-dimensional image data; and

generate the at least one structure based on the linear condensations and the rotated 3-dimensional image data.