MAPPING MULTIPLE VIEWS TO AN IDENTITY

• US 20200134862A1
• Filed: 10/30/2018
• Published: 04/30/2020
• Est. Priority Date: 10/30/2018
• Status: Active Grant

• Current Assignee: NCR Voyix Corporation
• Original Assignee: NCR Corporation (NCR Voyix Corporation)
• Inventors: Zucker, Brent Vance, Lieberman, Adam Justin
• Granted Patent: US 10,991,119 B2
• Time in Patent Office: Days
• CPC Class Codes: G06Q 30/0639 Item locations
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 7/292 Multi-camera tracking
• CPC Class Codes: G06T 7/74 involving reference images or patches
• CPC Class Codes: G06T 7/97 Determining parameters from multiple pictures depth or shape recovery from multiple images G06T7/55; stereo camera calibration G06T7/85

Abstract

Disclosed are systems and methods for mapping multiple views to an identity. The systems and methods may include receiving a plurality of images that depict an object. Attributes associated with the object may be extracted from the plurality of images. An identity of the object may be determined based on processing the attributes.

Claims

1. A method comprising:

receiving a plurality of images that depicts an object;

extracting attributes associated with the object from the plurality of images;

determining, based on processing the attributes, an identity of the object.

2. The method of claim 1, wherein the object is an item for purchase located in a store or a person located inside the store.

3. The method of claim 1, wherein receiving the plurality of images includes receiving the plurality of images from a plurality of camera located inside a store.

4. The method of claim 3, wherein each of the plurality of cameras is oriented at a different angle to on another.

5. The method of claim 1, wherein extracting the attributes includes storing the attributes in an N dimensional vector.

6. The method of claim 1, wherein determining, based on processing the attributes, the identity of the object includes processing an N dimensional vector including the attributes.

7. The method of claim 1, wherein determining, based on processing the attributes, the identity of the object includes:

querying a K-nearest neighbors database with an N dimensional vector including the attributes; and

assigning the identity of the object to the object based on a feature vector stored in the K-nearest neighbors database having a shortest distance to the N dimensional vector.

8. The method of claim 7, further comprising calculating a Euclidean distance between the N dimensional vector and a plurality of feature vectors stored in the K-nearest neighbors database, the feature vector having the shortest distance to the N dimensional vector being one of the plurality of feature vectors.

9. The method of claim 8, wherein assigning the identity of the object to the object includes determining when the Euclidean distance is within a confidence interval.

10. The method of claim 1, further includes cropping at least one of the plurality of images before extracting the attributes.

11. A system comprising:

a plurality of cameras arranged to capture images within a store;

a server that includes a processor; and

a non-transitory computer-readable storage medium storing instructions that, when executed by the processor, cause the server to perform operations comprising;

receiving a plurality of images that depicts an object,

extracting attributes associated with the object from the plurality of images, and

determining, based on processing the attributes, an identity of the object.

12. The system of claim 11, wherein the object is an item for purchase located in a store or a person located inside the store.

13. The system of claim 11, wherein receiving the plurality of images includes receiving the plurality of images from a plurality of camera located inside a store.

14. The system of claim 13, wherein each of the plurality of cameras is oriented at a different angle to on another.

15. The system of claim 11, wherein extracting the attributes includes storing the attributes in an N dimensional vector.

16. The system of claim 11, wherein determining, based on processing the attributes, the identity of the object includes processing an N dimensional vector including the attributes.

17. The system of claim 11, wherein determining, based on processing the attributes, the identity of the object includes:

querying a K-nearest neighbors database with an N dimensional vector including the attributes; and

assigning the identity of the object to the object based on a feature vector stored in the K-nearest neighbors database having a shortest distance to the N dimensional vector.

18. The system of claim 17, further comprising calculating a Euclidean distance between the N dimensional vector and a plurality of feature vectors stored in the K-nearest neighbors database, the feature vector having the shortest distance to the N dimensional vector being one of the plurality of feature vectors.

19. The system of claim 18, wherein assigning the identity of the object to the object includes determining when the Euclidean distance is within a confidence interval.

20. The system of claim 11, wherein the plurality of cameras are arranged within a check-in area of the store.