Tracking persons in an automated-checkout store

1. A person-tracking method, comprising:

receiving, by a computer system, image data collected by a plurality of image sensors, the received image data comprising data associated with a person in an automated-checkout store;

receiving, by the computer system, vibration data collected by one or more vibration sensors located below a floor of the automated-checkout store, wherein the vibration data is generated in response to footsteps on the floor;

identifying, by the computer system based on the received image data, a plurality of features of the person;

extracting, by the computer system based on the identified features, data associated with the person from the received image data, wherein the extracted data correspond to a plurality of time periods; and

determining, by the computer system, a location of the person in the automated-checkout store in each of the time periods based on the extracted data corresponding to the time period and the vibration data, wherein the determining comprises aggregating the extracted data collected by different image sensors based at least in part on a location and a line of sight of each of the image sensors, and combining the extracted data with the vibration data.

2. The method of claim 1 , wherein the identifying comprises:

assigning an identifier to the person; and

storing the features of the person in association with the identifier.

3. The method of claim 1 , wherein the extracting comprises:

for one of the time periods, determining a location of the person in a preceding time period;

identifying one or more of the image sensors each having a field of view that encompasses the determined location; and

obtaining data collected by the identified image sensors that correspond to the time period.

4. The method of claim 3 , further comprising:

determining a movement path of the person based on the location of the person in each of the time periods;

determining that the person exits the field of view of one or more of the identified image sensors and enters the field of view of one or more other image sensors; and

determining a location of the person in a subsequent time period based at least in part on data collected by the one or more other image sensors.

5. The method of claim 1 , further comprising:

for one of the time periods, detecting a failure of locating the person in a preceding time period;

extracting data associated with the person from the received image data that correspond to the one of the time periods based on the identified features of the person;

identifying one or more of the image sensors that collected the extracted data; and

determining the location of the person based on data collected by the identified image sensors.

6. The method of claim 1 , wherein the extracted data comprises a plurality of images captured by the image sensors, and wherein the aggregating the extracted data comprises:

identifying one or more pixels corresponding to the person in each of the images;

determining a plurality of lines in a three-dimensional space, wherein teach of the lines is determined based on a position of one of the identified pixels in the image containing the pixel and the line of sight of the image sensor capturing the image; and

determining one or more intersection areas of the lines.

7. The method of claim 1 , further comprising:

processing the image data collected by each group of image sensors to obtain analysis results; and

aggregating the analysis results associated with the plurality of groups of image sensors.

8. The method of claim 1 , wherein the determining the location of the person in each of the time periods further comprises:

determining a plurality of possible locations of the person each being associated with a probability value; and

selecting one of the possible locations as the location of the person based at least in part on the probability values.

9. The method of claim 1 , wherein the determining the location of the person in each of the time periods further comprises:

determining a plurality of possible locations of the person; and

selecting one of the possible locations as the location of the person based at least in part on a previous movement path of the person.

10. The method of claim 1 , further comprising:

receiving, by the computer system, weight data collected by one or more weight sensors located below the floor of the automated-checkout store, wherein the weight data is generated in response to footsteps on the floor, and the determining, by the computer system, the location of the person in the automated-checkout store in each of the time periods further comprises:

determining, by the computer system, the location of the person in the automated-checkout store in each of the time periods based on the extracted data corresponding to the time period, the vibration data, and the weight data, and

combining the extracted data with the vibration data and the weight data.

11. The method of claim 1 , further comprising:

receiving, by the computer system, electrical data collected by one or more piezo film sensors located in the floor of the automated-checkout store, wherein the electrical data is generated in response to footsteps on the floor, and the determining, by the computer system, the location of the person in the automated-checkout store in each of the time periods further comprises:

determining, by the computer system, the location of the person in the automated-checkout store in each of the time periods based on the extracted data corresponding to the time period, the vibration data, and the electrical data, and

combining the extracted data with the vibration data and the electrical data.

12. A system for tracking a person in an automated-checkout store comprising a computer system, one or more vibration sensors, and a plurality of image sensors, the computer system comprising one or more processors and one or more non-transitory computer-readable storage media storing instructions executable by the one or more processors to cause the system to perform operations comprising:

receiving, by the computer system, image data collected by the image sensors, the received image data comprising data associated with a person in the automated-checkout store;

receiving, by the computer system, vibration data collected by the one or more vibration sensors located below a floor of the automated-checkout store, wherein the vibration data is generated in response to footsteps on the floor;

identifying, by the computer system based on the received image data, a plurality of features of the person;

extracting, by the computer system based on the identified features, data associated with the person from the received image data, wherein the extracted data correspond to a plurality of time periods; and

determining, by the computer system, a location of the person in the automated-checkout store in each of the time periods based on the extracted data corresponding to the time period and the vibration data, wherein the determining comprises aggregating the extracted data collected by different image sensors based at least in part on a location and a line of sight of each of the image sensors, and combining the extracted data with the vibration data.

13. The system of claim 12 , wherein the identifying, by the computer system based on the received image data, a plurality of features of the person, comprises:

assigning an identifier to the person; and

storing the plurality of features of the person in association with the identifier.

14. The system of claim 12 , wherein the extracting, by the computer system based on the identified features, comprises:

for one of the time periods, determining a location of the person in a preceding time period;

identifying one or more of the image sensors each having a field of view that encompasses the determined location; and

obtaining data collected by the identified image sensors that correspond to the time period.

15. The system of claim 14 , wherein the operations further comprise:

determining a movement path of the person based on the location of the person in each of the time periods;

determining that the person exits the field of view of one or more of the identified image sensors and enters the field of view of one or more other image sensors; and

determining a location of the person in a subsequent time period based at least in part on data collected by the one or more other image sensors.

16. The system of claim 12 , wherein the operations further comprise:

for one of the time periods, detecting a failure of locating the person in a preceding time period;

extracting data associated with the person from the received image data that correspond to the one of the time periods based on the identified features of the person;

identifying one or more of the image sensors that collected the extracted data; and

determining the location of the person based on data collected by the identified image sensors.

17. The system of claim 12 , wherein the extracted data comprises a plurality of images captured by the image sensors, and wherein the aggregating the extracted data comprises:

identifying one or more pixels corresponding to the person in each of the images;

determining a plurality of lines in a three-dimensional space, wherein teach of the lines is determined based on a position of one of the identified pixels in the image containing the pixel and the line of sight of the image sensor capturing the image; and

determining one or more intersection areas of the lines.

18. The system of claim 12 , wherein the plurality of image sensors correspond to a plurality of groups each comprising one or more of the image sensors, and wherein the aggregating the extracted data comprises:

processing the image data collected by each group of image sensors to obtain analysis results; and

aggregating the analysis results associated with the plurality of groups of image sensors.

19. The system of claim 12 , wherein the determining the location of the person in each of the time periods further comprises:

determining a plurality of possible locations of the person each being associated with a probability value; and

selecting one of the possible locations as the location of the person based at least in part on the probability values.

20. The system of claim 12 , wherein the determining the location of the person in each of the time periods further comprises:

determining a plurality of possible locations of the person; and

selecting one of the possible locations as the location of the person based at least in part on a previous movement path of the person.