US 12,294,809 B1
AI-powered cloud-native network video recorder (NVR)
Song Cao, San Jose, CA (US); and Tianhu Yang, San Jose, CA (US)
Assigned to AGI7 Inc., San Jose, CA (US)
Filed by AGI7 Inc., San Jose, CA (US)
Filed on Apr. 5, 2024, as Appl. No. 18/628,532.
Int. Cl. H04N 5/77 (2006.01); G06V 10/70 (2022.01); G06V 20/40 (2022.01); G06V 20/52 (2022.01); G11B 27/031 (2006.01); H04N 5/91 (2006.01); H04N 7/18 (2006.01); H04N 19/40 (2014.01)
CPC H04N 5/77 (2013.01) [G06V 10/70 (2022.01); G06V 20/44 (2022.01); G06V 20/52 (2022.01); G11B 27/031 (2013.01); H04N 5/91 (2013.01); H04N 7/181 (2013.01); H04N 19/40 (2014.11)] 17 Claims
OG exemplary drawing
 
1. A Network Video Recorder (NVR) apparatus, comprising:
a plurality of Power over Ethernet (PoE) ports;
one or more processors and one or more non-transitory computer-readable memories coupled with the one or more processors; and
a camera-integration module, an Artificial Intelligence (AI) module, and a cloud-access module deployed in the one or more non-transitory computer-readable memories and executable by the one or more processors, wherein:
the camera-integration module is configured to connect a plurality of cameras to the NVR through the plurality of PoE ports and receive video data collected by the plurality of cameras,
wherein the camera-integration module is manufacturer-agnostic and comprises a pool of communication protocols respectively corresponding to a plurality of camera manufacturers,
the camera-integration module is further configured to activate, from the pool of communication protocols, a communication protocol corresponding to a camera manufacturer for connecting to a camera from the camera manufacturer, and
in response to a camera missing a corresponding communication protocol in the pool of communication protocols, prompting an alert to an operator to update a firmware of the NVR;
the AI module is configured to process the video data and generate video clips of interest; and
the cloud-access module is configured to:
store the video clips of interest in a local video staging storage, wherein a first subset of the video clips with faces unrecognized by the AI module are assigned with a higher priority than a second subset of video clips with faces recognized by the AI module, and
upload the first subset of video clips to a cloud storage before uploading the second subset of video clips.