PERFORMANCE OF IMAGE RECOGNITION ALGORITHMS BY PRUNING FEATURES, IMAGE SCALING, AND SPATIALLY CONSTRAINED FEATURE MATCHING
First Claim
1. A method for feature matching in image recognition, comprising:
- obtaining a plurality of keypoints for a query image;
defining a set of keypoint clusters for the query image based on keypoint distribution;
establishing anchor matches for the query image by matching at least a subset of keypoints for the query image to keypoints for one or more target images; and
matching a query keypoint cluster to a target keypoint cluster based on a subset of the anchor matches.
1 Assignment
0 Petitions
Accused Products
Abstract
A method for feature matching in image recognition is provided. First, image scaling may be based on a feature distribution across scale spaces for an image to estimate image size/resolution, where peak(s) in the keypoint distribution at different scales is used to track a dominant image scale and roughly track object sizes. Second, instead of using all detected features in an image for feature matching, keypoints may be pruned based on cluster density and/or the scale level in which the keypoints are detected. Keypoints falling within high-density clusters may be preferred over features falling within lower density clusters for purposes of feature matching. Third, inlier-to-outlier keypoint ratios are increased by spatially constraining keypoints into clusters in order to reduce or avoid geometric consistency checking for the image.
-
Citations
35 Claims
-
1. A method for feature matching in image recognition, comprising:
-
obtaining a plurality of keypoints for a query image; defining a set of keypoint clusters for the query image based on keypoint distribution; establishing anchor matches for the query image by matching at least a subset of keypoints for the query image to keypoints for one or more target images; and matching a query keypoint cluster to a target keypoint cluster based on a subset of the anchor matches. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
-
24. A feature matching device, comprising:
-
a storage device storing information for a plurality of target images; a processing circuit coupled to the storage device, the processing circuit adapted to; obtain a plurality of keypoints for a query image; define a set of keypoint clusters for the query image based on keypoint distribution; establish anchor matches for the query image by matching at least a subset of keypoints for the query image to keypoints for one or more target images; and match a query keypoint cluster to a target keypoint cluster based on a subset of the anchor matches. - View Dependent Claims (25, 26, 27, 28, 29, 30, 31, 32, 33)
-
-
34. A feature matching device, comprising:
-
means for obtaining a plurality of keypoints for a query image; means for defining a set of keypoint clusters for the query image based on keypoint distribution; means for establishing anchor matches for the query image by matching at least a subset of keypoints for the query image to keypoints for one or more target images; and means for matching a query keypoint cluster to a target keypoint cluster based on a subset of the anchor matches.
-
-
35. A processor-readable medium comprising one or more instructions operational on a feature matching device, which when executed by a processing circuit, causes the processing circuit to:
-
obtain a plurality of keypoints for a query image; define a set of keypoint clusters for the query image based on keypoint distribution; establish anchor matches for the query image by matching at least a subset of keypoints for the query image to keypoints for one or more target images; and match a query keypoint cluster to a target keypoint cluster based on a subset of the anchor matches.
-
Specification