IP Library › Granted Patent US 11,314,968
Granted Patent B2
US 11,314,968 · App. 16/646,215 · Granted Apr 26, 2022

Information processing apparatus, control method, and program

Inventor: Kazuyuki Sasaki (Tokyo, JP)
Assignee: NEC CORPORATION
G06K9/00348G06K9/00369G06K9/00771
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,314,968
App. No.
16/646,215
Filed
Mar 11, 2020
Granted
Apr 26, 2022
Kind
B2
Art Unit
2661
USPC
382/103
Abstract

An information processing apparatus ( 2000 ) detects leaving of an object to be determined from an object queue ( 20 ) included in video data ( 12 ) generated by a camera ( 10 ). The information processing apparatus ( 2000 ) infers a global behavior of the object to be determined on the basis of a relationship between the object to be determined and another object ( 22 ) included in the object queue ( 20 ). The information processing apparatus ( 2000 ) determines whether or not the object to be determined has left the object queue ( 20 ) on the basis of the global behavior of the object to be determined.

Claims (54)

1. An information processing system comprising:

at least one memory configured to store one or more instructions; and

at least one processor configured to execute the one or more instructions to:

detect a first object and a second object included in an object queue using a captured image in which the object queue is included, infer a behavior of the first object and a behavior of the second object, and infer a behavior of the first object related to the object sequence on the basis of the inferred behavior of the first object and the inferred behavior of the second object; and

determine whether or not the first object has left the object queue on the basis of the inferred behavior of the first object performed with respect to the object sequence.

2. The information processing system according to claim 1 ,

wherein the at least one processor is further configured to compute a likelihood of the first object having left the object queue on the basis of at least one of a distance between the first object and the second object, a difference in velocity between the first object and the second object, and an angle between a movement direction of the first object and a movement direction of the second object, and the inferred behavior of the first object, and determine whether or not the first object has left the object queue on the basis of the computed likelihood.

3. The information processing system according to claim 1 ,

wherein, for each of the first object and the second object, the at least one processor is further configured to infer a local behavior that is a behavior focused on an individual object using the captured image, and

infer the behavior of the first object related to the object sequence on the basis of a combination of the local behavior of the first object and the local behavior of the second object.

4. The information processing system according to claim 3 ,

wherein the local behavior is any one of a proceeding state that represents proceeding, a receding state that represents receding, and a standstill state that represents a standstill.

5. The information processing system according to claim 1 ,

wherein the behavior of the first object related to the object sequence is any of a first behavior that represents proceeding in the object queue, a second behavior that represents receding in the object queue, a third behavior that represents a standstill, and a fourth behavior that represents leaving.

6. The information processing system according to claim 5 ,

wherein the at least one processor is further configured to compute a likelihood of the first object having left the object queue by inputting a parameter representing at least one of a distance between the first object and the second object, a difference in velocity between the first object and the second object, and an angle between a movement direction of the first object and a movement direction of the second object, and the behavior of the first object related to the object sequence into a determination model, and determine whether or not the first object has left the object queue on the basis of the computed likelihood, and

in a case where the parameter is fixed, the determination model computes the likelihood that is greater when the behavior of the first object related to the object sequence is the fourth behavior than when the behavior of the first object related to the object sequence is other than the fourth behavior.

7. A control method executed by a computer, the method comprising:

detecting a first object and a second object included in an object queue using a captured image in which the object queue is included, inferring a behavior of the first object and a behavior of the second object, and infer a behavior of the first object related to the object sequence on the basis of the inferred behavior of the first object and the inferred behavior of the second object; and

determining whether or not the first object has left the object queue on the basis of the inferred behavior of the first object performed with respect to the object sequence.

8. The control method according to claim 7 ,

wherein a likelihood of the first object having left the object queue is computed on the basis of at least one of a distance between the first object and the second object, a difference in velocity between the first object and the second object, and an angle between a movement direction of the first object and a movement direction of the second object, and the inferred behavior of the first object related to the object sequence, and whether or not the first object has left the object queue is determined on the basis of the computed likelihood.

9. The control method according to claim 7 ,

wherein for each of the first object and the second object, a local behavior that is a behavior focused on an individual object is inferred using the captured image, and

the behavior of the first object related to the object sequence is inferred on the basis of a combination of the local behavior of the first object and the local behavior of the second object.

10. A non-transitory computer-readable storage medium storing a program causing a computer to execute:

the control method according to claim 7 .

11. An information processing system comprising:

at least one memory configured to store one or more instructions; and

at least one processor configured to execute the one or more instructions to:

infer a global behavior based on a relationship with a second object and a third object included in an object queue for a first object included in a captured image in which the object queue is included using the captured image; and

determine whether or not the first object cuts in the object queue on the basis of the inferred global behavior of the first object.

12. The information processing system according to claim 11 ,

wherein the at least one processor is further configured to compute a likelihood of the first object cutting in the object queue on the basis of at least one of a distance between the first object and the second object, a distance between the first object and the third object, an angle between a movement direction of the first object and a movement direction of the second object, a time-series change of the movement direction of the first object, and a staying time of the first object, and the inferred global behavior of the first object, and determine whether or not the first object cuts in the object queue on the basis of the computed likelihood.

13. The information processing system according to claim 11 ,

wherein, for each of the second object and the third object, the at least one processor is further configured to infer a local behavior that is a behavior focused on an individual object using the captured image, and

infer the global behavior of the first object on the basis of a combination of the local behavior of the second object and the local behavior of the third object.

14. The information processing system according to claim 13 ,

wherein the local behavior is any one of a proceeding state that represents proceeding, a receding state that represents receding, and a standstill state that represents a standstill.

15. The information processing system according to claim 11 ,

wherein the global behavior is any of a first behavior that represents proceeding in the object queue, a second behavior that represents receding in the object queue, a third behavior that represents a standstill, and a fourth behavior that represents cutting-in.

16. The information processing system according to claim 15 ,

wherein the at least one processor is further configured to compute a likelihood of the first object cutting in the object queue by inputting a parameter representing at least one of a distance between the first object and the second object, a distance between the first object and the third object, an angle between a movement direction of the first object and a movement direction of the second object, a time-series change of the movement direction of the first object, and a staying time of the first object, and the global behavior of the first object into a determination model, and determine whether or not the first object cuts in the object queue on the basis of the computed likelihood, and

in a case where the parameter is fixed, the determination model computes the likelihood that is greater when the global behavior of the first object is the fourth behavior than when the global behavior of the first object is other than the fourth behavior.

17. A control method executed by a computer, the method comprising:

inferring a global behavior based on a relationship with a second object and a third object included in an object queue for a first object included in a captured image in which the object queue is included using the captured image; and

determining whether or not the first object cuts in the object queue on the basis of the inferred global behavior of the first object.

18. The control method according to claim 17 ,

wherein a likelihood of the first object cutting in the object queue is computed on the basis of at least one of a distance between the first object and the second object, a distance between the first object and the third object, an angle between a movement direction of the first object and a movement direction of the second object, a time-series change of the movement direction of the first object, and a staying time of the first object, and the inferred global behavior of the first object, and whether or not the first object cuts in the object queue is determined on the basis of the computed likelihood.

19. The control method according to claim 17 ,

wherein for each of the second object and the third object, a local behavior that is a behavior focused on an individual object is inferred using the captured image, and

the global behavior of the first object is inferred on the basis of a combination of the local behavior of the second object and the local behavior of the third object.

20. A non-transitory computer-readable storage medium storing a program causing a computer to execute:

the control method according to claim 17 .

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 11, 2020
From: SASAKI, KAZUYUKI
To: NEC CORPORATION
Reel/Frame 052079/0548 →
Continuity (1)
Related Publication 20200279103A1 · Sep 3, 2020