| CPC G06V 20/44 (2022.01) [G06V 20/52 (2022.01); G06V 40/20 (2022.01); H04N 7/181 (2013.01)] | 18 Claims |
|
1. A method for identifying and monitoring an event of interest via an internet protocol video camera network installed at a location comprising:
a. providing the internet protocol (“IP”) video camera network operational at the location, wherein the IP video camera network comprises a plurality of passive IP enabled video cameras each installed at a plurality of different areas at the location, and wherein each of the plurality of passive IP enabled video cameras are:
i. in communications engagement with a graphics processing (“GPU”) server operational with a visual artificial intelligence engine; and
ii. configured with a first field of view (“FOV”) at a first time, thereby generating expected scene information for each camera;
b. determining at a second time whether there has been a substantial change in the first FOV of a first camera in the IP video camera network by:
i. identifying the first FOV for the first camera, wherein the first camera is positioned to acquire imaging information for a scene;
ii. acquiring imaging information from the first camera at the first time and at the second time;
iii. selecting one or more reference elements present in the scene from the imaging information acquired at the first time;
iv. processing the acquired imaging information to decompose the one or more reference elements obtained into polygons, wherein the one or more reference elements are present in the imaging information acquired of the scene at both the first and second times;
v. locating each of the polygons in the scene at the first and second times;
vi. comparing location information for the one or more reference elements at the first and second times;
vii. calculating a change in a location for each of the reference elements from positioning of each of the polygons in the scene at the first and second times, thereby determining whether the first FOV for the first camera has changed from the first to the second time,
and, if a change in the first FOV for the first camera has substantially changed from the first time to the second time, adjusting the first camera's FOV to substantially align with the first FOV;
c. detecting an event of interest for monitoring at the location, wherein the event of interest is associated with an activity of one or more humans at the location, and wherein the event of interest is detected by:
i. acquiring video stream information comprising at least two video streams, wherein each video stream is derived from a different video camera in the IP video camera network;
ii. transmitting the video stream information to the GPU server; and
iii. processing the video stream information via the visual artificial intelligence engine;
d. acquiring additional video stream information associated with the detected event of interest from the second time to a subsequent time;
e. transmitting the additional video stream information to the GPU server;
f. processing the additional video stream information via the visual artificial intelligence engine;
g. generating information associated with the detected event of interest; and
h. configuring the generated information for use in one or more of:
i. a report or a dashboard for use by an owner or manager of the location;
ii. a notification to one or more persons associated with the event of interest; or
iii. subsequent event detection processes.
|