Sign in to save

Bookmark this page so you can find it later.

Sign in to save

Bookmark this page so you can find it later.

Face unlock is an engineering system that lets a phone recognize its owner by measuring the shape and features of a face. Instead of only taking a normal photo, many advanced phones project invisible infrared light and sense how it reflects back. This matters because it combines optics, sensors, computer vision, and security in a device small enough to fit in a pocket.

The goal is fast access for the user while making it difficult for someone else to trick the system.

Key Facts

  • Infrared light used in face unlock is usually invisible to human eyes but detectable by phone sensors.
  • A dot projector can place thousands of tiny IR dots on a face to measure its 3D shape.
  • Depth can be estimated from geometry: distance = speed of light x time / 2 for time-of-flight systems.
  • A neural network compares the live face scan with a stored mathematical face model, not just a simple picture.
  • 3D face unlock can work in the dark because it brings its own infrared light source.
  • Security systems often use a match score, and unlock happens only if score >= threshold.

Vocabulary

Infrared light
Infrared light is electromagnetic radiation with wavelengths longer than visible red light, often used by sensors because people cannot see it.
Dot projector
A dot projector is a tiny device that shines a pattern of infrared dots onto a face so a camera can measure surface shape.
Depth map
A depth map is a data image in which each point stores how far that part of the face is from the camera.
Neural network
A neural network is a computer model that learns patterns from data and can compare a new face scan to a stored face model.
Biometric security
Biometric security uses body features such as a face, fingerprint, or iris to identify a person.

Common Mistakes to Avoid

  • Thinking face unlock is just a selfie is wrong because secure systems often use infrared sensors and 3D depth data, not only a color image.
  • Assuming a printed photo can always fool face unlock is wrong because depth sensing and liveness checks can detect that a flat picture lacks 3D structure.
  • Believing the phone stores your exact face photo is wrong because many systems store a mathematical template or model that is used for comparison.
  • Ignoring lighting conditions is a mistake because normal cameras need visible light, while infrared-based systems can work in darkness by projecting their own light.

Practice Questions

  1. 1 A face unlock sensor projects 30,000 infrared dots over a face. If 24,000 dots are clearly detected, what percent of the projected dots were detected?
  2. 2 A time-of-flight sensor measures that infrared light takes 4.0 x 10^-9 s to travel to a point on a face and return. Using speed of light = 3.0 x 10^8 m/s, how far away is that point?
  3. 3 Explain why a 3D infrared depth map makes face unlock harder to fool with a flat photograph than a normal camera image would.