Granted Patent B2

US 12,518,529 · App. 17/987,557 · Granted Jan 6, 2026

Inventors: Hugo Latapie (Long Beach, CA); Ozkan Kilic (Long Beach, CA); Adam James Lawrence (Pasadena, CA); Gaowen Liu (Austin, TX); Ramana Rao V. R. Kompella (Cupertino, CA)

Assignee: Cisco Technology, Inc.

G06V20/40G06V20/52G06V40/10G06V2201/08

Abstract

Claims (49)

1 . A method comprising:

tracking, by a device, objects in video data captured by one or more cameras in a location;

computing, by the device, spatial characteristics of the objects;

generating, by the device, a timeseries of the spatial characteristics of the objects over time, wherein the timeseries represents behaviors of the objects;

associating, by the device, different portions of the timeseries with behavioral regimes of the objects, wherein each behavioral regime is defined by a distinctive pattern in the timeseries indicative of an activity performed by the objects;

computing, by the device, a correlation between each behavioral regime of different objects based on their respective timeseries; and

making, by the device and based on the correlation, a determination that one of the objects is a reflection of another of the objects.

2 . The method as in claim 1 , wherein the spatial characteristics comprise detected centroids of the objects.

3 . The method as in claim 1 , wherein the behavioral regimes comprise different patterns of the timeseries associated with the objects performing various actions.

4 . The method as in claim 1 , wherein the objects are people in the location.

5 . The method as in claim 1 , further comprising:

identifying, by the device and based on the determination, a region of the video data as depicting a reflective surface in the location; and

stopping, by the device, tracking objects in the region.

6 . The method as in claim 5 , wherein the device identifies the region in part by:

computing a heatmap of reflections detected in the region.

7 . The method as in claim 1 , wherein the determination is based in part on an amount of mutual information between the timeseries.

8 . The method as in claim 1 , wherein the objects comprise vehicles.

9 . The method as in claim 1 , further comprising:

providing, by the device, an indication of the determination to a user interface for presentation to a user.

10 . The method as in claim 1 , wherein the device is an edge device in a network.

11 . An apparatus, comprising:

a network interface to communicate with a computer network;

a processor coupled to the network interface and configured to execute one or more processes; and

a memory configured to store a process that is executed by the processor, the process when executed configured to:

track objects in video data captured by one or more cameras in a location;

compute spatial characteristics of the objects;

generate a timeseries of the spatial characteristics of the objects over time, wherein the timeseries represents behaviors of the objects;

associate different portions of the timeseries with behavioral regimes of the objects, wherein each behavioral regime is defined by a distinctive pattern in the timeseries indicative of an activity performed by the objects;

compute a correlation between each behavioral regime of different objects based on their respective timeseries; and

make a determination, based on the correlation, that one of the objects is a reflection of another of the objects.

12 . The apparatus as in claim 11 , wherein the spatial characteristics comprise detected centroids of the objects.

13 . The apparatus as in claim 11 , wherein the behavioral regimes comprise different patterns of the timeseries associated with the objects performing various actions.

14 . The apparatus as in claim 11 , wherein the objects are people in the location.

15 . The apparatus as in claim 11 , wherein the process when executed is further configured to:

identify, based on the determination, a region of the video data as depicting a reflective surface in the location; and

stop tracking objects in the region.

16 . The apparatus as in claim 15 , wherein the apparatus identifies the region in part by:

computing a heatmap of reflections detected in the region.

17 . The apparatus as in claim 11 , wherein the determination is based in part on an amount of mutual information between the timeseries.

18 . The apparatus as in claim 11 , wherein the objects comprise vehicles.

19 . The apparatus as in claim 11 , wherein the process when executed is further configured to:

provide an indication of the determination to a user interface for presentation to a user.

20 . A tangible, non-transitory, computer-readable medium storing program instructions that cause a device to execute a process comprising:

tracking, by the device, objects in video data captured by one or more cameras in a location;

computing, by the device, spatial characteristics of the objects;

generating, by the device, a timeseries of the spatial characteristics of the objects over time, wherein the timeseries represents behaviors of the objects;

associating, by the device, different portions of the timeseries with behavioral regimes of the objects, wherein each behavioral regime is defined by a distinctive pattern in the timeseries indicative of an activity performed by the objects;

computing, by the device, a correlation between each behavioral regime of different objects based on their respective timeseries; and

making, by the device and based on the correlation, a determination that one of the objects is a reflection of another of the objects.

Assignments (1)

Continuity (1)

References Cited (109)