Systems and methods for broadcasting an audio or visual alert that includes a description of features of an ambient object extracted from an image captured by a camera of a doorbell device

Abstract

Systems and methods for broadcasting an audio or visual alert that includes a description of features associated with a person extracted from an image captured by a camera are provided. Such systems and methods can include the camera capturing the image when the person is within a field of view of the camera and a processor receiving the image from the camera, processing the image with an artificial intelligence model to identify and extract details of the features associated with the person, and initiating a broadcast of the audio or visual alert by an alert device associated with the camera, wherein the audio or visual alert can include the description of the features associated with the person.

1 - 20 . (canceled) 
     
     
         21 . A device comprising:
 a camera that, when activated, captures an image of a person within a field of view of the camera; and   a processor that (1) receives the image from the camera, (2) processes the image with an artificial intelligence model to identify and extract details of one or more visual features associated with the person, and (3) initiate a broadcast of an audio alert by an alert device associated with the camera,   wherein the audio alert includes an audio recitation of the details of the one or more visual features associated with the person.   
     
     
         22 . The device of  claim 21  wherein the alert device includes an audio device within a region outside of which the device is located. 
     
     
         23 . The device of  claim 21  wherein the alert device includes a mobile device of an authorized user of a region outside of which the device is located. 
     
     
         24 . The device of  claim 21  further comprising:
 a proximity detector that activates the camera when the person is detected within a detection range of the proximity detector. 
 
     
     
         25 . The device of  claim 21  wherein the processor processes the image to determine whether the person is a known object or a generic object, and wherein the processor initiates the audio or visual alert in response to classifying the person as the generic object. 
     
     
         26 . The device of  claim 21 , wherein the one or more visual features associated with the person comprise one or more of:
 a height of the person,   an eye color of the person,   a description of clothing worn by the person,   a physical build of the person, or   one or more textual indicators on the person or clothing worn by the person.   
     
     
         27 . The device of  claim 21  further comprising a transceiver, wherein the processor initiates the broadcast of the audio alert by the transceiver and to, and for output by, the alert device associated with the camera. 
     
     
         28 . A system comprising:
 a device;   a camera of the device that, when activated, captures an image of a person within a field of view of the camera; and   a cloud server remote from the device that (1) receives the image from the camera, (2) processes the image with an artificial intelligence model to identify and extract details of one or more visual features associated with the person, and (3) initiates a broadcast of an audio alert by an alert device associated with the camera,   wherein the audio alert includes an audio recitation of the details of the one or more visual features associated with the person.   
     
     
         29 . The system of  claim 28  wherein the alert device includes an audio device within a region outside of which the device is located. 
     
     
         30 . The system of  claim 28  wherein the alert device includes a mobile device of an authorized user of a region outside of which the device is located. 
     
     
         31 . The system of  claim 28  further comprising:
 a proximity detector that activates the camera when the person is detected within a detection range of the proximity detector. 
 
     
     
         32 . The system of  claim 28  wherein the cloud server processes the image to determine whether the person is a known object or a generic object, and wherein the cloud server initiates the audio or visual alert in response to classifying the person as the generic object. 
     
     
         33 . A method comprising:
 when a camera of a device is activated, capturing an image of a person within a field of view of the camera;   processing the image with an artificial intelligence model to identify and extract details of one or more visual features associated with the person; and   initiating a broadcast of an audio alert by an alert device associated with the camera,   wherein the audio alert includes an audio recitation of the details of the one or more visual features associated with the person.   
     
     
         34 . The method of  claim 33  wherein the alert device includes an audio device within a region outside of which the device is located. 
     
     
         35 . The method of  claim 33  wherein the alert device includes a mobile device of an authorized user of a region outside of which the device is located. 
     
     
         36 . The method of  claim 33  further comprising:
 activating the camera when the person is detected within a detection range of a proximity detector. 
 
     
     
         37 . The method of  claim 33  further comprising:
 determining whether the person is a known object or a generic object; and 
 in response to determining that the person is the generic object, initiating the broadcast of the audio or visual alert. 
 
     
     
         38 . The method of  claim 33  further comprising:
 processing the image with the artificial intelligence model at the device. 
 
     
     
         39 . The method of  claim 33  further comprising:
 processing the image with the artificial intelligence model at a cloud server that is remote from the device.