Meta platforms technologies, llc (20240233363). HIGH ACCURACY PEOPLE IDENTIFICATION OVER TIME BY LEVERAGING RE-IDENTIFICATION simplified abstract

HIGH ACCURACY PEOPLE IDENTIFICATION OVER TIME BY LEVERAGING RE-IDENTIFICATION

Organization Name

meta platforms technologies, llc

Inventor(s)

Mahdi Salmani Rahimi of San Francisco CA (US)

Rahul Nallamothu of Redwood City CA (US)

Samuel Franklin Pepose of Palo Alto CA (US)

HIGH ACCURACY PEOPLE IDENTIFICATION OVER TIME BY LEVERAGING RE-IDENTIFICATION - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240233363 titled 'HIGH ACCURACY PEOPLE IDENTIFICATION OVER TIME BY LEVERAGING RE-IDENTIFICATION

The method described in the abstract involves using machine learning models to authenticate participants in a video environment based on appearance information captured by a camera.

• The method determines the location of participants in an environment and detects appearance information of specific body regions.
• It calculates a confidence score to match the appearance information with pre-registered participant profiles.
• The confidence score is updated based on additional appearance information detected in subsequent frames.
• If the updated confidence score is above a threshold, the participant is authenticated.

1. 1. 1. Potential Applications:

This technology could be used for secure access control systems, video conferencing platforms, virtual event management, and surveillance systems.

1. 1. 1. Problems Solved:

This method addresses the need for accurate participant authentication in video environments, enhancing security and user verification processes.

1. 1. 1. Benefits:

- Improved security measures - Enhanced user verification - Streamlined access control processes

1. 1. 1. Commercial Applications:

The technology can be applied in industries such as cybersecurity, video communication software development, event management, and surveillance technology.

1. 1. 1. Questions about Participant Authentication:
         1. 1. How does this method improve upon traditional authentication processes?

This method utilizes appearance information and machine learning models to authenticate participants, providing a more secure and efficient process compared to traditional methods.

1. 1. 1. 1. 2. What are the potential privacy concerns associated with using this technology?

Privacy concerns may arise regarding the collection and storage of appearance information for participant authentication. It is essential to address these concerns through transparent data handling practices and consent mechanisms.

Original Abstract Submitted

in one embodiment, a method, by one or more computing systems, includes determining, based on frames captured by a camera, a plurality of participants are located in an environment, locating, within a first frame, a first body region of a first participant of the plurality of participants, detecting, at a first time, appearance information of the first body region of the first participant, calculating, using one or more machine-learning models, a confidence score corresponding to a match between the appearance information of the first participant at the first time and one or more profiles of pre-registered participants, updating, using the one or more machine-learning models, the confidence score based on one or more additional appearance information detected within additional frames, determining whether the updated confidence score is above a predetermined threshold, and in response to determining the updated confidence score is above the predetermined threshold, authenticating the first participant.