Calibration of Wi-Fi localization from video localization
In some implementations, video camera networks can be used to track people or objects and determine their locations. In some implementations, location accuracy better than one meter can be achieved using video cameras. According to some implementations, Wi-Fi localization and video localization systems can be fused together to perform calibration. In some implementations, by using the video network to continuously update the Wi-Fi localization calibration database, Wi-Fi location accuracy can be improved.
1. A method comprising:
obtaining, from a transmitter associated with a movable target, wireless signal pattern data at a plurality of access points corresponding to a plurality of fixed locations, the wireless signal pattern data including a signal strength measured by the plurality of access points at the plurality of fixed locations;
determining a wireless feature vector associated with the movable target, the wireless feature vector being determined based on the wireless signal pattern data;
obtaining video data associated with the movable target;
calculating occupancy probabilities of the movable target based on the video data;
estimating a location of the movable target based on the wireless feature vector and the occupancy probabilities; and
updating calibration data for the plurality of access points corresponding to the plurality of fixed locations based on the estimated location of the movable target.
2. The method of claim 1 , wherein obtaining the wireless signal pattern data is based on receiving wireless signals.
3. The method of claim 1 ,
wherein the calibration data is further updated based on the wireless signal feature vector.
4. The method of claim 1 , wherein estimating the location of the movable target further comprises:
determining a first location of an object based on the video data;
determining a second location of the transmitter based on the wireless signal pattern data obtained from the transmitter; and
determining, based on the first location and the second location, that the object corresponds to the transmitter.
5. The method of claim 1 , wherein estimating the location of the movable target further comprises:
tracking movement of an object along a first path based on the video data;
tracking movement of the transmitter along a second path based on the wireless signal pattern data; and
determining, based on the first path and the second path, that the object corresponds to the transmitter.
6. The method of claim 1 , wherein estimating the location of the movable target further comprises:
receiving video data from a plurality of cameras having overlapping fields of view, the video data including video of an object in the overlapping fields of view; and
determining a three-dimensional position of the object based on the video data received from the plurality of cameras.
7. The method of claim 1 , wherein the wireless signal pattern data identifies each of the plurality of fixed locations.
8. A system comprising:
one or more processors; and
a computer-readable medium including one or more sequences of instructions which, when executed by the one or more processors, causes:
obtaining, from a transmitter associated with a movable target, wireless signal pattern data at a plurality of access points corresponding to a plurality of fixed locations from a transmitter associated with a movable target, the wireless signal pattern data including a signal strength measured by the plurality of access points at the plurality of fixed locations;
determining a wireless feature vector associated with the movable target, the wireless feature vector being determined based on the wireless signal pattern data;
obtaining video data associated with the movable target;
determining occupancy probabilities of the movable target based on the video data;
estimating a location of the movable target based on the wireless feature vector and the occupancy probabilities; and
updating calibration data for the plurality of access points corresponding to the plurality of fixed locations based on the estimated location of the movable target.
9. The system of claim 8 , wherein the instructions for obtaining the wireless signal pattern data is based on receiving wireless signals.
10. The system of claim 8 , wherein
the calibration data is further updated based on the wireless signal feature vector.
11. The system of claim 8 , wherein the instructions that cause estimating the location of the target further comprise instructions that cause:
determining a first location of an object based on the video data;
determining a second location of the transmitter based on the wireless signal pattern data obtained from the transmitter; and
determining, based on the first location and the second location, that the object corresponds to the transmitter.
12. The system of claim 8 , wherein the instructions that cause estimating the location of the movable target further comprise instructions that cause:
tracking movement of an object along a first path based on the video data;
tracking movement of the transmitter along a second path based on the wireless signal pattern data; and
determining, based on the first path and the second path, that the object corresponds to the transmitter.
13. The system of claim 8 , wherein the instructions that cause estimating the location of the movable target further comprise instructions that cause:
receiving video data from a plurality of cameras having overlapping fields of view, the video data including video of an object in the overlapping fields of view; and
determining a three-dimensional position of the object based on the video data received from the plurality of cameras.
14. The system of claim 8 , wherein the wireless signal pattern data identifies each of the plurality of fixed locations.
15. A non-transitory computer-readable medium including one or more sequences of instructions which, when executed by one or more processors, causes:
obtaining, from a transmitter associated with a movable target, wireless signal pattern data at a plurality of access points corresponding to a plurality of fixed locations, the wireless signal pattern data including a signal strength measured by the plurality of access points at the plurality of fixed locations;
determining a wireless feature vector associated with the movable target, the wireless feature vector being determined based on the wireless signal pattern data;
obtaining video data associated with the movable target;
calculating occupancy probabilities of the movable target based on the video data;
estimating a location of the movable target based on the wireless feature vector and the occupancy probabilities; and
updating calibration data for the plurality of access points corresponding to the plurality of fixed locations based on the estimated location of the movable target.
16. The non-transitory computer-readable medium of claim 15 , wherein the instructions for obtaining the wireless signal pattern data is based on receiving wireless signals.
17. The non-transitory computer-readable medium of claim 15 , wherein the instructions further cause:
updating the wireless signal pattern data based on the wireless signal feature vector.
18. The non-transitory computer-readable medium of claim 15 , wherein the instructions that cause estimating the location of the movable target comprise instructions that cause:
determining a first location of an object based on the video data;
determining a second location of the wireless transmitter based on the wireless signals received from the wireless transmitter; and
determining, based on the first location and the second location, that the object corresponds to the wireless transmitter.
19. The non-transitory computer-readable medium of claim 15 , wherein the instructions that cause estimating the location of the movable target comprise instructions that cause:
tracking movement of an object along a first path based on the video data;
tracking movement of the wireless transmitter along a second path based on the wireless signals received from the wireless transmitter; and
determining, based on the first path and the second path, that the object corresponds to the wireless transmitter.
20. The non-transitory computer-readable medium of claim 15 , wherein the instructions that cause determining occupancy probabilities of the movable target based on the video data comprise instructions that cause:
receiving the video data from a plurality of cameras having overlapping fields of view, the video data including video of an object in the overlapping fields of view; and
determining a three-dimensional position of the movable object based on the video data.
21. The non-transitory computer-readable medium of claim 15 , wherein the wireless signal pattern data identifies each of the plurality of fixed locations.