Facial recognition IOT application for multi-stream video using forecasting tools
One example method includes accessing an appearance history of a person, the appearance history including information concerning an appearance of a person at a particular time, generating, based on the appearance history, a forecast that comprises a probability that the person will appear again at some future point in time, determining that the probability meets or exceeds a threshold, and updating a high probability group database to include a facial image of a face of the person.
1. A method, comprising:
accessing an appearance history of a person, the appearance history including information concerning an appearance of a person at a particular time;
dividing a time period into segments;
collecting time series information for the time period;
calculating, using the time series information, a probability that the person will appear again at another particular point in time, and the another particular point in time falls within one of the segments;
generating, based on the appearance history, and in real time, a forecast that comprises the probability that the person will physically appear again at the another particular time and particular location;
determining that the probability meets or exceeds a threshold; and
updating a high probability group database to include a facial image of a face of the person.
2. The method as recited in claim 1 , wherein the appearance history comprises one or more timestamped facial images of the person.
3. The method as recited in claim 1 , further comprising collecting a plurality of video streams, and generating the appearance history of the person using information from the plurality of video streams.
4. The method as recited in claim 1 , further comprising collecting a plurality of video streams, and the plurality of video streams comprises a plurality of facial images of the person, each of the facial images comprising a different respective perspective of the face of the person.
5. The method as recited in claim 1 , further comprising capturing, from a plurality of different angles, a respective plurality of facial images of the person.
6. The method as recited in claim 1 , wherein, at the time the forecast is generated, the person has not yet appeared at the another particular time and particular location.
7. The method as recited in claim 1 , wherein updating of the high probability group database to include a facial image of a face of the person is performed before the another particular point in time that the person was forecast to appear.
8. The method as recited in claim 1 , wherein all faces in the high probability group database collectively form a subset of faces included in a database of known faces.
9. The method as recited in claim 1 , wherein the facial image added to the high probability group database is obtained from a video stream.
10. A non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations comprising:
accessing an appearance history of a person, the appearance history including information concerning an appearance of a person at a particular time;
dividing a time period into segments;
collecting time series information for the time period;
calculating, using the time series information, a probability that the person will appear again at another particular point in time, and the another particular point in time falls within one of the segments;
generating, based on the appearance history, and in real time, a forecast that comprises the probability that the person will physically appear again at the another particular time and particular location;
determining that the probability meets or exceeds a threshold; and
updating a high probability group database to include a facial image of a face of the person.
11. The non-transitory storage medium as recited in claim 10 , wherein the appearance history comprises one or more timestamped facial images of the person.
12. The non-transitory storage medium as recited in claim 10 , wherein the operations further comprise collecting a plurality of video streams, and generating the appearance history of the person using information from the plurality of video streams.
13. The non-transitory storage medium as recited in claim 10 , wherein the operations further comprise collecting a plurality of video streams, and the plurality of video streams comprises a plurality of facial images of the person, each of the facial images comprising a different respective perspective of the face of the person.
14. The non-transitory storage medium as recited in claim 10 , wherein the operations further comprise capturing, from a plurality of different angles, a respective plurality of facial images of the person.
15. The non-transitory storage medium as recited in claim 10 , wherein, at the time the forecast is generated, the person has not yet appeared at the another particular time and particular location.
16. The non-transitory storage medium as recited in claim 10 , wherein updating of the high probability group database to include a facial image of a face of the person is performed before the another particular point in time that the person was forecast to appear.
17. The non-transitory storage medium as recited in claim 10 , wherein all faces in the high probability group database collectively form a subset of faces included in a database of known faces.
18. The non-transitory storage medium as recited in claim 10 , wherein the facial image added to the high probability group database is obtained from a video stream.