Large scale image recognition using global signatures and local feature information
First Claim
1. A computer-based method for conducting an image recognition search, comprising:
- obtaining, by a computing device, one or more global signatures for a query image, wherein a global signature is a full image descriptor that can represent an entire image, and wherein the one or more global signatures includes a machine learning signature;
determining, by the computing device, a ranking order for a plurality of document images based on nearest neighbor relations between document signatures corresponding to the plurality of document images and each one of the one or more global signatures for the query image;
selecting, by the computing device, a subset of the plurality of document images based on the determined ranking order;
obtaining, by the computing device, additional document data corresponding to the selected subset of the plurality of document images, wherein the obtained additional document data comprises, for each document image of the selected subset of the plurality of document images, an at least partially compressed data set that includes a global signature of the document image and, for each local feature of the document image, one or more of (1) an indication of at least one of a location, orientation and scale, and (2) an indication of at least one of a 3D location and a surface normal of the 3D location; and
generating, by the computing device, a search result of document images filtered by using a geometric verification between the additional document data corresponding to the selected subset of the plurality of document images and the query image, wherein the geometric verification, using a distance check threshold, compares at least a portion of the at least partially compressed data set for each document image of the selected subset of the plurality of document images with a feature descriptor from the query image.
4 Assignments
0 Petitions
Accused Products
Abstract
Techniques are provided that include receiving one or more global signatures for a query image in response to an image recognition query, wherein some of the plurality of global signatures are generated using local descriptors corresponding to different cropped versions of the image. A ranking order is determined for a plurality of document images based on nearest neighbor relations between document signatures corresponding to the plurality of document images and each one of the one or more global signatures for the query image. A subset of the plurality of document images is selected based on the determined ranking order. Additional document data corresponding to the selected subset of the plurality of document images is obtained, and a search result is generated based on a geometric verification between the additional document data corresponding to the selected subset of the plurality of document images and the query image.
-
Citations
38 Claims
-
1. A computer-based method for conducting an image recognition search, comprising:
-
obtaining, by a computing device, one or more global signatures for a query image, wherein a global signature is a full image descriptor that can represent an entire image, and wherein the one or more global signatures includes a machine learning signature; determining, by the computing device, a ranking order for a plurality of document images based on nearest neighbor relations between document signatures corresponding to the plurality of document images and each one of the one or more global signatures for the query image; selecting, by the computing device, a subset of the plurality of document images based on the determined ranking order; obtaining, by the computing device, additional document data corresponding to the selected subset of the plurality of document images, wherein the obtained additional document data comprises, for each document image of the selected subset of the plurality of document images, an at least partially compressed data set that includes a global signature of the document image and, for each local feature of the document image, one or more of (1) an indication of at least one of a location, orientation and scale, and (2) an indication of at least one of a 3D location and a surface normal of the 3D location; and generating, by the computing device, a search result of document images filtered by using a geometric verification between the additional document data corresponding to the selected subset of the plurality of document images and the query image, wherein the geometric verification, using a distance check threshold, compares at least a portion of the at least partially compressed data set for each document image of the selected subset of the plurality of document images with a feature descriptor from the query image. - 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, 25, 26, 27, 28, 29)
-
-
30. A system for conducting an image recognition search, comprising:
-
a processor; a main memory device; a persistent storage device; and a recognition engine executable on the processor according to software instructions stored in at least one of the main memory device and the persistent storage device, the software instructions, when executed, causing the processor to; obtain one or more global signatures for a query image, wherein a global signature is a full image descriptor that can represent an entire image, and wherein the one or more global signatures includes a machine learning signature; determine a ranking order for the plurality of document images based on nearest neighbor relations between the document signatures corresponding to a plurality of document images from the main memory device and each one of the one or more global signatures for the query image; select a subset of the plurality of document images based on the determined ranking order, load into the main memory device additional document data corresponding to the selected subset of the plurality of document images from the persistent storage device, wherein the loaded additional document data comprises, for each document image of the selected subset of the plurality of document images, an at least partially compressed data set that includes a global signature of the document image and, for each local feature of the document image, one or more of (1) an indication of at least one of a location, orientation and scale, and (2) an indication of at least one of a 3D location and a surface normal of the 3D location; and generate a search result of document images filtered by using a geometric verification between the additional document data and the query image, wherein the geometric verification, using a distance check threshold, compares at least a portion of the at least partially compressed data set for each document image of the selected subset of the plurality of document images with a feature descriptor from the query image. - View Dependent Claims (31, 32)
-
-
33. A computer program product embedded in a non-transitory computer readable medium comprising instructions executable by a computer processor for conducting an image recognition search, which, when executed by a processor, cause the processor to perform one or more steps comprising:
-
obtaining one or more global signatures for a query image, wherein a global signature is a full image descriptor that can represent an entire image, and wherein the one or more global signatures includes a machine learning signature; determining a ranking order for a plurality of document images based on nearest neighbor relations between document signatures corresponding to the plurality of document images and each one of the one or more global signatures for the query image; selecting a subset of the plurality of document images based on the determined ranking order; obtaining additional document data corresponding to the selected subset of the plurality of document images, wherein the obtained additional document data comprises, for each document image of the selected subset of the plurality of document images, an at least partially compressed data set that includes a global signature of the document image and, for each local feature of the document image, one or more of (1) an indication of at least one of a location, orientation and scale, and (2) an indication of at least one of a 3D location and a surface normal of the 3D location; and generating a search result of document images filtered by using a geometric verification between the additional document data corresponding to the selected subset of the plurality of document images and the query image, wherein the geometric verification, using a distance check threshold, compares at least a portion of the at least partially compressed data set for each document image of the selected subset of the plurality of document images with a feature descriptor from the query image. - View Dependent Claims (34, 35)
-
-
36. A system for conducting an image recognition search, comprising:
-
means for obtaining one or more global signatures for a query image, wherein a global signature is a full image descriptor that can represent an entire image, and wherein the one or more global signatures includes a machine learning signature; means for determining a ranking order for a plurality of document images based on nearest neighbor relations between document signatures corresponding to the plurality of document images and each one of the one or more global signatures for the query image; means for selecting a subset of the plurality of document images based on the determined ranking order; means for obtaining additional document data corresponding to the selected subset of the plurality of document images, wherein the obtained additional document data comprises, for each document image of the selected subset of the plurality of document images, an at least partially compressed data set that includes a global signature of the document image and, for each local feature of the document image, one or more of (1) an indication of at least one of a location, orientation and scale, and (2) an indication of at least one of a 3D location and a surface normal of the 3D location; and means for generating a search result of document images filtered by using a geometric verification between the additional document data and the query image, wherein the geometric verification using a distance check threshold compares at least a portion of the at least partially compressed data set for each document image of the selected subset of the plurality of document images with a feature descriptor from the query image. - View Dependent Claims (37, 38)
-
Specification