Monitoring target object by multiple cameras

1. A method for monitoring a target object by multiple cameras, the method comprising:

dynamically building a surveillance network based on attributes from a plurality of cameras, the surveillance network defining reachability by the target object between locations of the plurality of cameras; and

in response to determining that a target object is detected at a first camera of the plurality of cameras, monitoring the target object based on the reachability between locations of the-plurality of cameras, wherein monitoring, in response to determining that the target object is detected at a first camera among the plurality of cameras, the target object based on the reachability between locations of the plurality of cameras comprises:

searching among the plurality of cameras for at least one second camera that is in a reachable relationship with the first camera including determining one or more cameras that are unreachable from the location of the first camera based upon, at least in part, the reachability between the location of the first camera and each other camera of the plurality of cameras; and

detecting the target object at the at least one second camera by searching for the target object in images collected from the at least one second camera.

2. The method according to claim 1 , wherein building a surveillance network based on attributes of the plurality of cameras comprises:

acquiring location information of the plurality of cameras and path information between the plurality of cameras, the path information defining reachability by the target object between locations of the plurality of cameras; and

building the surveillance network based on the location information and the path information.

3. The method according to claim 2 , wherein building the surveillance network based on the location information and the path information comprises:

using camera attribute information; and

setting weights of edges in the surveillance network based on the path information.

4. The method according to claim 1 , wherein determining that a target object is detected at a first camera among the plurality of cameras comprises:

extracting metadata from content monitored at the plurality of cameras, wherein the metadata comprises at least one of: vehicle information and person information; and

in response to determining that the metadata matches the target object, determining that the target object is detected at the first camera among the plurality of cameras.

5. The method according to claim 1 , wherein detecting the target object at the at least one second camera comprises:

detecting the target object at the at least one second camera within a predefined time period.

6. The method according to claim 1 , wherein detecting the target object at the at least one second camera comprises:

in response to the target object being detected at a third camera among the at least one second camera, prompting a movement path of the target object as “first camera→third camera”; and

identifying the third camera as the first camera.

7. The method according claim 1 , wherein a camera among the plurality of cameras is one of: a fixed camera and a mobile camera.

8. An apparatus for monitoring a target object by multiple cameras, the apparatus configured to dynamically build a surveillance network based on attributes from a plurality of cameras, the surveillance network defining reachability by the target object between locations of the plurality of cameras; and

in response to determining that a target object is detected at a first camera among the plurality of cameras, monitor the target object based on the reachability between locations of the plurality of cameras, wherein monitoring, in response to determining that the target object is detected at a first camera among the plurality of cameras, the target object based on the reachability between locations of the plurality of cameras comprises:

searching among the plurality of cameras for at least one second camera that is in a reachable relationship with the first camera including determining one or more cameras that are unreachable from the location of the first camera based upon, at least in part, the reachability between the location of the first camera and each other camera of the plurality of cameras; and

detecting the target object at the at least one second camera by searching for the target object in images collected from the at least one second camera.

9. The apparatus according to claim 8 , the apparatus further configured to:

acquire location information of the plurality of cameras and path information between the plurality of cameras, the path information defining reachability by the target object between locations of the plurality of cameras; and

build the surveillance network based on the location information and the path information.

10. The apparatus according to claim 9 , further configured to:

use camera attribute information to build the surveillance network; and

set weights of edges in the surveillance network based on the path information.

11. The apparatus according claim 8 , further configured to:

extract metadata from content monitored at the plurality of cameras, wherein the metadata comprises at least one of: vehicle information and person information; and

in response to determining that the metadata matches the target object, determine that the target object is detected at the first camera among the plurality of cameras.

12. The apparatus according to claim 8 , further configured to:

detect the target object at the at least one second camera within a predefined time period.

13. The apparatus according to claim 8 , further configured to:

in response to the target object being detected at a third camera among the at least one second camera, prompt a movement path of the target object as “first camera→third camera”; and

identify the third camera as the first camera.

14. The apparatus according to claim 8 , wherein a camera among the plurality of cameras is one of: a fixed camera and a mobile camera.

15. A computer program product comprising:

a non-transitory computer readable medium encoded with computer executable program code, wherein the code enables execution across one or more processors for:

dynamically building a surveillance network based on attributes from a plurality of cameras, the surveillance network defining reachability by the target object between locations of the plurality of cameras; and

in response to determining that a target object is detected at a first camera among the plurality of cameras, monitoring the target object based on the reachability between locations of the plurality of cameras, wherein monitoring the target object based on the reachability between locations of the plurality of cameras includes:

searching among the plurality of cameras for at least one second camera that is in a reachable relationship with the first camera including determining one or more cameras that are unreachable from the location of the first camera based upon, at least in part, the reachability between the location of the first camera and each other camera of the plurality of cameras; and

detecting the target object at the at least one second camera by searching for the target object in images collected from the at least one second camera.

16. The computer program product according to claim 15 , wherein building a surveillance network based on attributes of the plurality of cameras comprises:

acquiring location information of the plurality of cameras and path information between the plurality of cameras, the path information defining reachability by the target object between locations of the plurality of cameras; and

building the surveillance network based on the location information and the path information, further comprising:

using camera attribute information to build the surveillance network; and

setting weights of edges in the surveillance network based on the path information.

17. The computer program product according to claim 16 , wherein determining that a target object is detected at a first camera among the plurality of cameras comprises:

extracting metadata from content monitored at the plurality of cameras, wherein the metadata comprises at least one of: vehicle information and person information; and

in response to determining that the metadata matches the target object, determining that the target object is detected at the first camera among the plurality of cameras.

18. The computer program product according to claim 15 , wherein detecting the target object at the at least one second camera comprises:

in response to the target object being detected at a third camera among the at least one second camera, prompting a movement path of the target object as “first camera→third camera”; and

identifying the third camera as the first camera.