Grouping images by location

1. A method of matching background regions between a reference image and a target image, comprising

(a) determining the background regions for the reference image and the target image;

(b) locating feature points in the background of the reference and target images;

(c) matching feature points in the reference and target images;

(d) clustering matched points in the target image into spatially compact regions in an image coordinate space to produce a set of target clusters;

(e) clustering matched points in the reference image into spatially compact regions in the image coordinate space to produce a set of reference clusters;

(f) finding a matching target cluster for each of the reference clusters by counting a number of matched points between target and reference clusters;

(g) removing reference clusters based on the count of the number of matched points;

(h) computing a match score between the reference and target images from the remaining reference clusters; and

(i) using the match score to determine whether the reference and target images are captured in the same location.

2. The method of claim 1 , wherein step (h) includes:

(i) removing points in the reference clusters that are not coherent with a global trajectory of matched points; and

(ii) removing points in the reference clusters that violate compactness of clusters.

3. The method of claim 1 , wherein the match score in step (h) incorporates:

geographic distance between reference and target images.

4. The method of claim 1 , wherein a label that indicates a common location is provided for the reference and target images when it is determined that the reference and target images are captured at the same location.