Systems and methods for identifying events within video content using intelligent search query
Inventors
Janakiraman, Kirupakar • Muthusamy, Baskaran • Patel, Keyurbhai
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-12639370-B2
Publication Date
2026-05-26
Expiration Date
Abstract
A video management system (VMS) may search for one or more objects and/or events in one or more video streams, and may receive time-stamped metadata that may identify one or more objects and/or events occurring in the corresponding video stream as well as an identifier that uniquely identifies the corresponding video stream. A user may enter a query into a video query engine, wherein the video query engine includes one or more cognitive models. The VMS may apply the search query to the time-stamped metadata via the video query engine to search for one or more objects and/or events in the one or more video streams that match the search query, and returning a search result to the user.
Core Innovation
A security surveillance system includes a plurality of remote sites and a central hub operatively coupled to the plurality of remote sites and a workstation. Each remote site receives a video stream from each of a plurality of video cameras at the corresponding remote site and generates time-stamped metadata for each of the video streams captured by the plurality of video cameras. The time-stamped metadata identifies objects and/or events in the corresponding video stream.
The central hub receives the time-stamped metadata from the video cameras at each remote site. The central hub receives a user search query from a security operator entered via a user interface of the workstation and applies a search query based at least in part on the user search query to identify one or more matching results. Each matching result includes a site identifier, a camera identifier, and a time stamp, or includes a camera identifier and a time stamp.
The workstation receives the one or more matching results from the central hub and displays a listing of the one or more matching results on the user interface. The listing displays the site identifier, the camera identifier, and the time stamp, or displays the camera identifier, and includes a link for each matching result. In response to selection of a link for a particular matching result, the workstation requests and receives a video clip of the matching result from the remote site as captured by the corresponding video camera, and displays the received video clip on the user interface.
Claims Coverage
The partial content includes three independent claims (clm-00001, clm-00010, clm-00018). Across these independent claims, the core claim coverage centers on generating time-stamped metadata identifying objects and/or events per video stream at remote sites, applying a user search query to that metadata at a central hub to identify matching results that include camera identifier(s) and time stamp, and using link-driven interaction from the workstation to request and display corresponding video clips.
Generating time-stamped metadata identifying objects and/or events
A plurality of remote sites receive a video stream from each of a plurality of video cameras at the corresponding remote site and generate time-stamped metadata for each video stream, wherein the time-stamped metadata identifies objects and/or events in the corresponding video stream.
Applying a user search query to time-stamped metadata to identify matching results
A central hub receives the time-stamped metadata from the plurality of video cameras and receives a user search query from a security operator via the workstation user interface, and applies a search query based at least in part on the user search query to identify one or more matching results.
Matching results include identifiers and a time stamp
Each matching result includes a site identifier identifying the remote site and a camera identifier identifying the video camera corresponding to the matching result, and a time stamp identifying a time of the matching result, or includes the camera identifier and the time stamp.
Displaying a listing with a link that triggers clip retrieval
The workstation receives the one or more matching results from the central hub and displays a listing on the user interface; the listing displays the identifiers and time stamp and includes a link for each matching result, wherein in response to selection of the link the workstation requests and receives a video clip of the matching result from the remote site as captured by the corresponding video camera and displays the received video clip.
Method for identifying objects or events using metadata search and clip display
A method receives video streams at a remote site, generates time-stamped metadata identifying objects and/or events, receives the metadata at a central hub, receives a user search query from a security operator via a workstation user interface, applies a search query to identify one or more matching results including a camera identifier and a time stamp, displays a listing with links at the workstation, receives selection of a link, and in response requests and receives a video clip of the matching result from the remote site and displays the received video clip.
The independent claims collectively cover a security surveillance workflow in which remote sites generate time-stamped metadata identifying objects and/or events, a central hub applies a security-operator user search query to that metadata to identify matching results with camera identifier(s) and time stamp, and the workstation displays the results in a listing with links that trigger request, receipt, and display of corresponding video clips.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?