Automatic Object Detection and Recognition via a Camera System

Abstract

Systems and methods for automatically detecting and recognizing human faces by a smart camera system are described herein. A camera located in a physical space may detect a triggering event and begin recording video and/or audio for a predetermined time period. The camera also processes the video to determine if any human faces are present. If so, the camera updates the metadata file associated with the recorded video with information identifying specific video frames that contain a human face. The recorded video and metadata file are transmitted to video analysis software, which performs further facial analysis on the selected video frames, in an attempt to identify the person(s) detected in the video. Results of the facial recognition process are presented on a user computing device.

What is claimed is: 
     
         1 . A system for facial detection and recognition via a camera system, the system comprising:
 a camera comprising:
 a lens; 
 at least one sensor; 
 a microphone; 
 a first processor and a first memory, configured to:
 detect a triggering event in a physical space around the camera; 
 record video for a predetermined time period, in response to detecting the triggering event; 
 execute firmware on the camera to detect that at least one human face is present in at least one video frame of the recorded video; 
 update a metadata file associated with the recorded video with information regarding the at least one video frame that has the detected at least one human face; and 
 transmit the recorded video and the updated metadata file to a video analysis system in communication with the camera; and 
 
   the video analysis system comprising:
 a second processor and second memory, the second processor configured to:
 receive the transmitted recorded video and updated metadata file from the camera; 
 extract from the recorded video the at least one video frame with the detected at least one human face, based on information in the updated metadata file; 
 perform facial recognition analysis on the extracted at least one video frame to identify the detected at least one human face; 
 transmit and present a result of the identified detected at least one human face to a user interface on a user computing device in communication with the video analysis system. 
 
   
     
     
         2 . The camera system of  claim 1 , wherein the recorded video further comprises audio. 
     
     
         3 . The camera system of  claim 1 , wherein the detected triggering event is a loud noise in the physical space around the camera that is detected by the microphone. 
     
     
         4 . The camera system of  claim 1 , wherein the detected triggering event is a movement in the physical space around the camera that is detected by the at least one sensor. 
     
     
         5 . The camera system of  claim 1 , wherein the predetermined time period for recording video in response to detecting the triggering event is less than one minute. 
     
     
         6 . The camera system of  claim 1 , wherein the detecting that at least one human face is present in at least one video frame of the recorded video occurs substantially simultaneously with recording the video in response to detecting the triggering event. 
     
     
         7 . The camera system of  claim 1 , wherein the first processor is a specialized video processor. 
     
     
         8 . The camera system of  claim 1 , wherein the video analysis system further comprises a database communicatively coupled to the second processor, the database storing previously identified human faces and associated identify information. 
     
     
         9 . The camera system of  claim 1 , wherein the result of the facial recognition analysis is a name of the detected at least one human face. 
     
     
         10 . The camera system of  claim 1 , wherein the result of the facial recognition analysis is an unknown identity of the detected at least one human face. 
     
     
         11 . A method for facial detection and recognition via a camera system, the method comprising:
 detecting, by a camera, a triggering event in a physical space around the camera;   recording video for a predetermined time period by the camera, in response to detecting the triggering event;   executing firmware on the camera to detect that at least one human face is present in at least one video frame of the recorded video;   updating, by the camera, a metadata file associated with the recorded video with information regarding the at least one video frame that has the detected at least one human face;   transmitting the recorded video and the updated metadata file, by the camera to a video analysis system in communication with the camera;   receiving the recorded video and updated metadata file by the video analysis system;   extracting the at least one video frame with the detected at least one human face from the recorded video, based on information in the updated metadata file;   performing facial recognition analysis on the extracted at least one video frame, to identify the detected at least one human face; and   transmit and present a result of the identified detected at least one human face to a user interface on a user computing device in communication with the video analysis system.   
     
     
         12 . The method of  claim 11 , wherein the detected triggering event is a loud noise in the physical space around the camera that is detected by the microphone. 
     
     
         13 . The method of  claim 11 , wherein the detected triggering event is a movement in the physical space around the camera that is detected by the at least one sensor. 
     
     
         14 . The method of  claim 11 , wherein the predetermined time period for recording video in response to detecting the triggering event is less than one minute. 
     
     
         15 . The method of  claim 11 , wherein the detecting that at least one human face is present in at least one video frame of the recorded video occurs substantially simultaneously with recording the video in response to detecting the triggering event. 
     
     
         16 . The method of  claim 11 , wherein the first processor is a specialized video processor. 
     
     
         17 . The method of  claim 11 , wherein the video analysis system further comprises a database communicatively coupled to the second processor, the database storing previously identified human faces and associated identify information. 
     
     
         18 . The method of  claim 11 , wherein the result of the facial recognition analysis is a name of the detected at least one human face. 
     
     
         19 . The method of  claim 11 , wherein the result of the facial recognition analysis is an unknown identity of the detected at least one human face. 
     
     
         20 . A system for object detection and recognition via a camera system, the system comprising:
 a camera comprising:
 a lens; 
 at least one sensor; 
 a microphone; 
 a first processor and a first memory, configured to:
 detect a triggering event in a physical space around the camera; 
 record video for a predetermined time period, in response to detecting the triggering event; 
 execute firmware on the camera to detect that at least one known object is present in at least one video frame of the recorded video; 
 update a metadata file associated with the recorded video with information regarding the at least one video frame that has the detected at least one known object; and 
 transmit the recorded video and the updated metadata file to a video analysis system in communication with the camera; and 
 
   the video analysis system comprising:
 a second processor and second memory, the second processor configured to:
 receive the transmitted recorded video and updated metadata file from the camera; 
 extract from the recorded video the at least one video frame with the detected at least one object, based on information in the updated metadata file; 
 perform further recognition analysis on the extracted at least one video frame to identify the detected at least one known object; 
 transmit and present a result of the identified detected at least one known object to a user interface on a user computing device in communication with the video analysis system.