Methods and systems for object monitoring

1. A system, comprising:

at least one hardware processor to:

detect a target object in first image data provided by a first camera;

determine a first plurality of neighboring cameras of the first camera that have detected the target object;

determine at least one of the first plurality of neighboring cameras as a next node of the first camera;

divide a first field of view of the first camera into a plurality of regions;

determine that the target object departs from the first field of view via a first region of the plurality of regions;

determine, based on the first region, a motion direction of the target object in the first field of view of the first camera; and

generate, based on the motion direction, first topological data for the first camera and the next node of the first camera.

2. The system of claim 1 , wherein each of the first plurality of neighboring cameras is connected to the first camera via a threshold number of intermediate devices.

3. The system of claim 1 , wherein, to determine at least one of the first plurality of neighboring cameras as the next node of the first camera, the hardware processor is further to:

acquire a first plurality of time instants corresponding to the target object entering a plurality of fields of view of the first plurality of neighboring cameras;

select, from the first plurality of time instants, a first time instant;

determine a first neighboring camera of the first plurality of neighboring cameras corresponding to the first time instant; and

determine the first neighboring camera as the next node of the first camera.

4. The system of claim 3 , wherein the hardware processor is further to:

determine that a second time instant is the same as a third time instant, wherein the first plurality of time instants include the second time instant and the third time instant; and

select, from the first plurality of neighboring cameras, a second plurality of neighboring cameras based on the second time instant and the third time instant.

5. The system of claim 4 , wherein the hardware processor is further to:

acquire a second plurality of time instants corresponding to departure of the target object from a second plurality of fields of view of each of the second plurality of neighboring cameras;

select, from the second plurality of time instants, a fourth time instant; and

determine the next node of the first camera based on the fourth time instant.

6. The system of claim 4 , wherein the hardware processor is further to:

detect at least one object based on image data provided by each of the second plurality of neighboring cameras;

compare at least one feature of the at least one detected object and at least one feature of the target object to determine similarities between the at least one detected object and the target object; and

determine the next node of the first camera based on the comparison.

7. The system of claim 1 , wherein, to determine a plurality of neighboring cameras of the first camera that have detected the target object, the hardware processor is further to:

acquire second image data provided by the first plurality of neighboring cameras; and

determine that the first plurality of neighboring cameras of the first camera have detected the target object based on the second image data.

8. The system of claim 1 , wherein the hardware processor is further to:

generate second topological data for a plurality of cameras based on the first topological data for the first camera and the next node of the first camera.

9. The system of claim 1 , wherein, to determine the first plurality of neighboring cameras of the first camera, the hardware processor is further to:

receive, from at least one of the first plurality of cameras, a response indicative of detection of the target object, wherein the response is responsive to a request for tracking the target object; and

determine the first plurality of neighboring cameras based on the response.

10. The system of claim 1 , wherein the hardware processor is further to:

send, to the next node of the first camera, a request for tracking the target object; and

receive, from the next node of the first camera, a response, wherein the response includes an indication of detection of the target object.

11. A method implemented on a computing device having at least one processor and at least one computer-readable storage medium, the method comprising:

detecting a target object in first image data provided by a first camera;

determining a first plurality of neighboring cameras of the first camera that have detected the target object;

determining at least one of the first plurality of neighboring cameras as a next node of the first camera;

dividing a first field of view of the first camera into a plurality of regions;

determining that the target object departs from the first field of view via a first region of the plurality of regions;

determining, based on the first region, a motion direction of the target object in the first field of view of the first camera; and

generating, based on the motion direction, first topological data for the first camera and the next node of the first camera.

12. The method of claim 11 , wherein each of the first plurality of neighboring cameras is connected to the first camera via a threshold number of intermediate devices.

13. The method of claim 11 , wherein the determining at least one of the first plurality of neighboring cameras as the next node of the first camera includes:

acquiring a first plurality of time instants corresponding to the target object entering a plurality of fields of view of the first plurality of neighboring cameras;

selecting, from the first plurality of time instants, a first time instant;

determining a first neighboring camera of the first plurality of neighboring cameras corresponding to the first time instant; and

determining the first neighboring camera as the next node of the first camera.

14. The method of claim 13 , further including:

determining that a second time instant is the same as a third time instant, wherein the first plurality of time instants include the second time instant and the third time instant; and

selecting, from the first plurality of neighboring cameras, a second plurality of neighboring cameras based on the second time instant and the third time instant.

15. The method of claim 14 , further including:

acquiring a second plurality of time instants corresponding to departure of the target object from a second plurality of fields of view of each of the second plurality of neighboring cameras;

selecting, from the second plurality of time instants, a fourth time instant; and

determining the next node of the first camera based on the fourth time instant.

16. A non-transitory computer readable medium, comprising at least one set of instructions, wherein when executed by at least one processor of a computing device, the at least one set of instructions causes the computing device to perform a method, the method comprising:

detecting a target object in first image data provided by a first camera;

determining a first plurality of neighboring cameras of the first camera that have detected the target object;

determining at least one of the first plurality of neighboring cameras as a next node of the first camera;

dividing a first field of view of the first camera into a plurality of regions;

determining that the target object departs from the first field of view via a first region of the plurality of regions;

determining, based on the first region, a motion direction of the target object in the first field of view of the first camera; and

generating, based on the motion direction, first topological data for the first camera and the next node of the first camera.