| CPC G06V 20/53 (2022.01) [G06T 7/246 (2017.01); G06T 7/248 (2017.01); G06T 7/254 (2017.01); G06V 40/10 (2022.01); G06T 2207/10016 (2013.01); G06T 2207/30196 (2013.01); G06T 2207/30242 (2013.01); G08G 1/01 (2013.01); G08G 1/04 (2013.01)] | 17 Claims |

|
1. A device comprising:
at least one memory storing instructions; and
at least one processor configured to execute the instructions to perform:
obtaining image frames of people, the image frames comprising a first image frame of an area and a second image frame of the area;
determining, by machine learning, a number of the people in a first region of the first image frame and a number of the people in a second region of the second image frame; and
estimating a direction in which at least some of the people are flowing based on the number of the people in the first region of the first image frame and the number of the people in the second region of the second image frame,
wherein the first region represents a first local region of the area,
wherein the second region represents a second local region of the area, and
wherein the first local region and the second local region are isolated from each other such that:
the first local region and the second local region neither overlap nor contact each other,
the first local region does not surround the second local region, and
the second local region does not surround the first local region,
wherein of local regions of the area, the local regions comprising at least the first local region and the second local region, none of the local regions overlaps any other of the local regions, none of all local regions contacts any other of the local regions, none of the local regions surrounds any other of the local regions, and
wherein the local regions of the area comprise all local regions of the area.
|