Entry prevention of persons of interest from venues and events using facial recognition
AssigneeWicket, LLC
InventorsSamson Timoner, Robert Banks, Kelly A. Bucey, Patrick L. Quinlan, Gennady Livitz, Sanjay Manandhar
Life of this patent
Estimating the term…
Abstract
A system uses facial recognition to exclude persons of interest (e.g., “undesirables”) from events and/or venues. Such a system can include a combination of cameras, edge processing devices, and servers that are in communications coupling (e.g., via a network) and that are on premise and/or in the cloud to recognize such persons of interest (POI) and interdict and prevent such POI from entering, traversing, and/or attending (collectively, “entering”) such venues or events.
Claims12
An edge computing-based system of detecting persons of interest, comprising:
- A. a server digital data processor that creates embeddings of facial images of each of a plurality of persons of interest and that downloads the embeddings to one or more edge computing devices local to a premises,
- B. one or more cameras, each of which acquires a stream of images of persons at the premises,
- C. one or more edge computing devices that are in communications coupling with the server digital data processor and the one or more cameras, each edge computing device processing a said stream of images to identify person of interest by converting each face in the stream of images to a vector, and finding a degree of similarity between each such vector and the downloaded embeddings,
- D. each edge computing device generating a match signal for a facial image for which a degree of similarity is above a threshold.
An edge computing-based method of detecting persons of interest, comprising:
- A. creating, with a server digital data processor, embeddings of facial images of each of a plurality of persons of interest,
- B. downloading, from the server digital data processor to an edge computing device local to a premises, embeddings of the persons of interest,
- C. acquiring, with a camera local to the premises, a stream of images of persons at the premises,
- D. processing, with the edge computing device, the stream of images to identify persons of interest, wherein the processing step includes
- i. converting each face in the stream of images to a vector, and
- ii. finding a degree of similarity between each such vector and the downloaded embeddings,
- E. generating a match signal for a facial image for which a degree of similarity is above a threshold.
Dependencies are inferred from the claim text.