STEM Sims Logo

Face Recognition

Face Recognition Thumbnail
Facial recognition software can analyze a face and come up with a match in a matter of seconds. Do you have what it takes to be able to determine if two different pictures are of the same face or not? Your challenge is to take measurements of different faces and determine which pictures are of the same person. If you feel ready, grab your ruler and get going!

Face Recognition Brochure

Face Recognition Brochure

How does facial recognition work?

Facial recognition software analyzes the distances between specific features of the face, called nodal points. The average human face has approximately 80 nodal points. Common measurements taken by facial recognition software include the shape of the cheekbones, depth of the eye sockets, and the distance between the eyes. The computer then searches for a match by comparing the data it has collected with data from other faces in its database.

Why use face recognition?

People, businesses, and the public sector use face recognition systems in a variety of ways. Some current uses include controlling access to buildings, identifying people with criminal backgrounds, and monitoring changes in a person's features during the treatment of certain diseases. One of the largest users of face recognition is the U.S. Department of State, which has over 75 million photographs they actively use for visa processing. In the near future, you may not need a credit card, voter card, or even a formal driver's license. Instead, your face will be your most important form of identification. The main goal of all face recognition systems is to "provide the right person with the right privileges the right access at the right time1." For other uses of face recognition systems, check out this article on using fuzzy clues to identify people and threats: . 1. Woodward, J., C. Horn, J. Gatune, and A. Thomas. Rand Public Safety and Justice. Biometrics: A look at facial recognition. 2003.

Are face recognition systems legal?

Legal Status Quo: We do not have a legal right of privacy in the facial features we show in public. "What a person knowingly exposes to the public. . . is not a subject of Fourth Amendment protection." United States v. Miller, 425 U.S. 435 (1976) "No person can have a reasonable expectation that others will not know the sound of his voice, any more than he can reasonably expect that his face will be a mystery to the world." United States v. Dionisio, 410 U.S. 1 (1973)

How accurate is facial recognition?

If the image quality is well controlled, facial recognition can identify someone accurately approximately 90% of the time - the same accuracy rate as identifying someone with fingerprints! However, if the quality of the images that the facial recognition software is analyzing are not carefully controlled and are not good quality, the accuracy rate drops significantly.

Walkthrough

Please wait while your video loads...
You need to log in to access this simulation.