Object linking method, object linking apparatus, and storage medium

1. An object linking method comprising:

detecting each of at least one first object based on a video image, each of the at least one first object having each first state that is one of states including a moving state and a stopping state;

detecting each of at least one second object based on each data measured by each inertial sensor in each of the at least one second object, each of the at least one second object having each second state that is one of states including a moving state and a stopping state; and

linking, by a computer, each of the at least one first object and each of the at least one second object in one-to-one, based on each first change of each first state and each second change of each second state, wherein

when a specified first object of the at least one first object is not linked to any of the at least one second object, the specified first object is linked to a virtual second object that is added to the at least one second object, and

when a specified second object of the at least one second object is not linked to any of the at least one first object, the specified second object is linked to a virtual first object that is added to the at least one first object.

2. The object linking method according to claim 1 , wherein

the linking of each of the at least one first object and each of the at least one second object in one-to-one is based on a square matrix, each row of the square matrix corresponding to each of the at least one first object to which the virtual first object is added when each column of the square matrix corresponding to each of the at least one second object to which the virtual second object is added, each column of the square matrix corresponding to each of the at least one first object to which the virtual first object is added when each row of the square matrix corresponding to each of the at least one second object to which the virtual second object is added.

3. The object linking method according to claim 2 , further comprising:

deleting, when a first object become undetected, the undetected first object and a second object or a virtual second object corresponding to the undetected first object from the square matrix, and

deleting, when a second object become undetected, the undetected second object and a first object or a virtual first object corresponding to the undetected second object from the square matrix.

4. The object linking method according to claim 1 , wherein

the linking of each of the at least one first object and each of the at least one second object in one-to-one is based on each certainty factor that is given to each of combinations of each of the at least one first object and each of the at least one second object based on each first change of each first state and each second change of each second state.

5. The object linking method according to claim 4 , wherein

each certainty factor of each of event pairs of each first event and each second event is calculated based on a specified probability model, each first event relating to each change of each first state of each first object of each of the event pairs, each second event relating to each change of each second state of each second object of each of the event pairs, each first event and each second event being detected within a specified time window.

6. The object linking method according to claim 5 , wherein

each certainty factor is calculated based on a first undetected event when a first event is undetected, and

each certainty factor is calculated based on a second undetected event when a second event is undetected.

7. The object linking method according to claim 5 , wherein

the specified probability model is based on each difference between each first time in which each first event is detected and each second time in which each second event is detected.

8. The object linking method according to claim 5 , wherein

the specified probability model is based on each difference between each first time in which each first event is detected and each second time in which each second event is detected, and

when one of a first event and a second event is undetected, a difference of the first event and the second event is obtained based on uniform distribution within the specified time window.

9. The object linking method according to claim 8 , wherein

the specified probability model includes each probability of each transition of each first event and each second event.

10. The object linking method according to claim 5 , wherein

the specified probability model estimates each of subsequent first events and each of subsequent second events based on each probability of each transition of each first event and each second event,

the subsequent first events include a first undetected event, and

the subsequent second events include a second undetected event.

11. The object linking method according to claim 5 , further comprising:

selecting a first combination of each of object pairs of each first object and each second object from combination candidates of each of the object pairs so that a first sum of certainty factors of the first combination of each of the object pairs is the greatest in the combination candidates, wherein

the linking of each of the at least one first object and each of the at least one second object in one-to-one is based on the first combination of each of object pairs.

12. The object linking method according to claim 11 , further comprising:

selecting a second combination of each of object pairs of each first object and each second object from combination candidates of each of the object pairs so that a second sum of certainty factors of the second combination of each of the object pairs is the second greatest in the combination candidates, wherein

the linking of each of the at least one first object and each of the at least one second object in one-to-one is based on the first combination of each of object pairs when a difference between the first sum and the second sum is greater than a specified value.

13. The object linking method according to claim 1 , wherein

each of the at least one first object is each person, and

each of the at least one second object is each device carried by each person.

14. An object linking apparatus comprising:

a memory; and

a processor coupled to the memory and configured to:

detect each of at least one first object based on a video image, each of the at least one first object having each first state that is one of states including a moving state and a stopping state,

detect each of at least one second object based on each data measured by each inertial sensor in each of the at least one second object, each of the at least one second object having each second state that is one of states including a moving state and a stopping state, and

link each of the at least one first object and each of the at least one second object in one-to-one, based on each first change of each first state and each second change of each second state, wherein

when a specified first object of the at least one first object is not linked to any of the at least one second object, the specified first object is linked to a virtual second object that is added to the at least one second object, and

when a specified second object of the at least one second object is not linked to any of the at least one first object, the specified second object is linked to a virtual first object that is added to the at least one first object.

15. A non-transitory computer-readable storage medium storing a program that causes a computer to execute a process, the computer including a memory, the process comprising:

detecting each of at least one first object based on a video image, each of the at least one first object having each first state that is one of states including a moving state and a stopping state;

detecting each of at least one second object based on each data measured by each inertial sensor in each of the at least one second object, each of the at least one second object having each second state that is one of states including a moving state and a stopping state; and

linking each of the at least one first object and each of the at least one second object in one-to-one, based on each first change of each first state and each second change of each second state, wherein

when a specified first object of the at least one first object is not linked to any of the at least one second object, the specified first object is linked to a virtual second object that is added to the at least one second object, and

when a specified second object of the at least one second object is not linked to any of the at least one first object, the specified second object is linked to a virtual first object that is added to the at least one first object.