3-D RECONSTRUCTION USING AUGMENTED REALITY FRAMEWORKS

1 . (canceled)

2 . A system for generating a scaling factor for a building structure in a 3-D scene, comprising:

one or more processors;

memory, wherein the memory stores one or more programs capable of execution by the one or more processors, and the one or more programs comprise instructions for:

obtaining a plurality of images of the building structure, wherein the plurality of images comprises image features co-visible among at least two images;

deriving reference poses for an imager based on the image features;

obtaining world map data including real-world poses of the imager associated with the plurality of images, wherein the real-world poses are derived from an augmented reality framework and comprise geometric position information;

selecting at least two candidate poses from the real-world poses; and

calculating a scaling factor of corresponding reference poses based on the geometric position information of the at least two candidate poses.

3 . The system of claim 2 , further comprising generating a digital 3-D representation for the building structure based on the plurality of images.

4 . The system of claim 3 , further comprising applying the scaling factor to the digital 3-D representation.

5 . The system of claim 4 , further comprising extracting a measurement between two pixels in the digital 3-D representation according to the scaling factor.

6 . The system of claim 2 wherein the world map data is obtained while capturing the plurality of images.

7 . The system of claim 6 , wherein the world map data includes tracking states for the imager for an associated real-world pose, and selecting the at least two candidate poses from the real-world poses comprises selecting real-world poses based on a tracking state validation.

8 . The system of claim 7 , wherein the tracking state validation is a high confidence position.

9 . The system of claim 2 , wherein selecting the candidate poses from the real-world poses is based on translation changes between the real-world poses proportional with translation changes of corresponding reference poses.

10 . The system of claim 2 , wherein selecting the candidate poses from the real-world poses is based on three-dimensional vector changes between the real-world poses consistent with three-dimensional vector changes of corresponding reference poses.

11 . The system of claim 2 , wherein selecting the candidate poses from the real-world poses comprises selecting from an inlier set of real-world poses.

12 . The system of claim 11 , wherein the inlier set of real-world poses is a subsample of real-world pose pairs with scaling factors within a statistical threshold of at least one other real-world pose pair.

13 . The system of claim 11 , wherein selecting from an inlier set comprises:

selecting two or more of the real-world poses and corresponding reference poses;

generating a respective scale factor based on the selected two or more real-world poses and corresponding reference poses; and

selecting the inlier set as pairs of two or more real-world poses and corresponding reference poses having respective scale factors within a statistical threshold of one another.

14 . The system of claim 13 , wherein the statistical threshold is a least mean squares analysis.

15 . The system of claim 13 , wherein the statistical threshold is a random sample consensus analysis.

16 . The system of claim 11 , wherein the inlier set of real-world poses is identified by:

pairing each real-world pose with at least one other real-world pose;

selecting a corresponding reference pose for each real-world pose of the pair;

generating a respective scaling factor of the building structure based on paired real-world poses and corresponding reference poses; and

identifying as the inlier set pairs that generate scaling factors within a statistical threshold of one another.

17 . The system of claim 16 , wherein the statistical threshold is a least mean squares analysis.

18 . The system of claim 16 wherein the statistical threshold is a random sample consensus analysis.

19 . The system of claim 2 , wherein calculating the scaling factor for the 3-D representation comprises:

establishing correspondence between the candidate poses and the reference poses;

identifying a first pose and a second pose of the candidate poses separated by a first distance;

identifying a third pose and a fourth pose of the reference poses separated by a second distance, the third pose and the fourth pose corresponding to the first pose and the second pose, respectively; and

computing the scaling factor as a ratio between the first distance and the second distance.

20 . The system of claim 2 , wherein selecting the candidate poses from the real-world poses comprises performing processing to identify a subset of the real-world poses.

21 . The system of claim 20 , wherein the processing to identify the subset of the real-world poses comprises evaluating potential ones of the real-world poses by comparing respective scaling factors for the building structure produced by pairs of the real-world poses and using a threshold to identify the subset of real-world poses used in generating the scaling factor.