Systems and methods for object recognition

• US 8,447,863 B1
• Filed: 05/07/2012
• Issued: 05/21/2013
• Est. Priority Date: 05/06/2011
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Francis, Anthony G. Jr., Poursohi, Arshan
• Primary Examiner(s): JEAN GILLES, JUDE
• Application Number:
  US13/465,453
• Time in Patent Office: 379 Days
• Field of Search: 709/226, 709/224, 709/201, 709/223, 709/227, 709/245
• US Class Current: 709/226
• CPC Class Codes: G06F 16/2468 Fuzzy queries
• CPC Class Codes: H04W 4/02 Services making use of location information
• CPC Class Codes: H04W 4/029 Location-based management or tracking services
• CPC Class Codes: H04W 4/60 Subscription-based services using application servers or record carriers, e.g. SIM application toolkits

Abstract

Examples disclose systems and methods for recognizing objects. A method may be executable to receive a query from a robot. The query may include identification data associated with an object and contextual data associated with the object. The query may also include situational data. The method may also be executable to identify the object based at least in part on the data in the query received from the robot. Further, the method may be executable to send data associated with the identified object to the robot in response to the query.

Claims

1. A method comprising:

receiving, from a robot having at least one sensor, a query comprising;

identification data obtained by the at least one sensor and associated with an unidentified object,

contextual data obtained by the at least one sensor and associated with a location of the unidentified object, and

situational data comprising a command obtained by the at least one sensor at a time proximate to the contextual data being obtained;

identifying a plurality of candidate objects based on the identification data;

making a determination of whether the unidentified object is an identified object in the plurality of candidate objects based at least in part on the contextual data and the situational data; and

sending data associated with the identified object to the robot based on the determination.

2. The method of claim 1, wherein the identification data is one of (i) information read from a code, a tag, or a beacon associated with the unidentified object, (ii) an image of the unidentified object, and (iii) a name of the unidentified object.

3. The method of claim 1, wherein the contextual data is further associated with at least one of (i) location data associated with the robot, and (ii) information associated with other objects within a predetermined distance of the unidentified object.

4. The method of claim 1, wherein the situational data further comprises action data associated with an action performed by the robot at the time proximate to the contextual data being obtained.

5. The method of claim 1, wherein each of the plurality of candidate objects has a probability of correctly corresponding to the unidentified object, the method further comprising:

selecting one candidate of the plurality of candidate objects as the identified object based on the selected candidate'"'"'s probability of correctly corresponding to the unidentified object.

6. The method of claim 1, wherein the data associated with the identified object includes at least one of (i) an identifier associated with the identified object, (ii) one or more physical characteristics of the identified object, and (iii) one or more instructions for interacting with the identified object.

7. The method of claim 1, further comprising:

receiving feedback associated with the identified object from the robot, wherein the feedback confirms or corrects the data associated with the identified object that was previously sent to the robot; and

updating the data associated with the identified object based on the feedback.

8. The method of claim 7, further comprising:

determining whether the identified object correctly corresponds to the unidentified object associated with the query received from the robot based on the feedback data; and

sending a new identified object to the robot in response to determining that the identified object does not correctly correspond to the unidentified object associated with the query received from the robot.

9. The method of claim 7, further comprising:

storing details of the query received from the robot, the identified object, and the feedback data in a history log associated with the identified object.

10. The method of claim 7, wherein the query from the robot is based on a command received by the robot from a user associated with the robot.