Back to results
US 12,293,583 B2Granted
Notification priority sequencing for video security
AssigneeArlo Technologies, Inc.
InventorsPeiman Amini, Joseph Amalan Arul Emmanuel
Life of this patent
Estimating the term…
Abstract
Determining a sequence for providing a notification regarding activity recorded by a camera is described. In one aspect, a priority sequence for can be determined based on a variety of characteristics of the available devices registered with the home security system of the camera.
Claims20
A camera comprising:
- at least one hardware processor;
- and
- at least one non-transitory memory storing instructions, which, when executed by the at least one hardware processor, cause the camera to:
- capture video at a particular time,
- wherein information describing the particular time is embedded within the video;
- extract a feature vector from the video,
- wherein the feature vector describes an event depicted by the video and the particular time;
- determine, using a machine learning model embedded in the camera, a priority sequence, based on the feature vector, for providing a notification to at least one of a plurality of user devices,
- wherein the machine learning model is trained based on user preferences and characteristics of the plurality of user devices,
- wherein the user preferences are related to the particular time and the characteristics of the plurality of user devices, and
- wherein determining the priority sequence is performed using edge computing;
- and
- provide the notification to the at least one of the plurality of user devices based on the priority sequence.
10 more elements
A computer-implemented method comprising:
- receiving, by a computer system, a video captured by a camera at a particular time,
- wherein information describing the particular time is embedded within the video;
- extracting a feature vector from the video,
- wherein the feature vector describes an event depicted by the video and the particular time;
- determining, using a machine learning model, a priority sequence, based on the feature vector, for providing a notification, based on the video, to at least one of a plurality of user devices,
- wherein the machine learning model is trained based on user preferences and characteristics of the plurality of user devices;
- and
- providing the notification to the at least one of the plurality of user devices based on the priority sequence.
5 more elements
A base station comprising:
- at least one non-transitory memory storing instructions, which, when executed by at least one hardware processor, cause the base station to:
- receive video, from a camera, at a particular time,
- wherein information identifying the camera and the particular time is embedded within the video;
- extract a feature vector from the video,
- wherein the feature vector describes the particular time;
- generate, using a machine learning model, an extended-reality version of the video based on the feature vector and user preferences;
- determine, using the machine learning model, a priority sequence, based on the feature vector, for providing the extended-reality version of the video to at least one of a plurality of user devices;
- and
- send the extended-reality version of the video to the at least one of the plurality of user devices based on the priority sequence.
6 more elements
Dependencies are inferred from the claim text.