YOLO (You Only Look Once)

← Back to the article

YOLO looks at one picture and quickly finds objects in it.

What it does

YOLO looks at one picture and quickly finds objects in it.

Bounding Box: Rectangle around an object • FPS: Frames per second

How it works

  • Look at the whole image once.
  • Draw boxes around objects it sees.
  • Name each object, such as car, person, or bike.

Uses and applications

  • A self-driving car can quickly notice people, cars, and traffic lights.
  • A security camera can count how many people entered a room.
  • A robot can find a toy or tool on a table.

In short

YOLO is fast because it looks once and finds objects right away.

Related methods

  • Faster R-CNN: A two-stage detector that proposes regions first, then checks them.
  • Mask R-CNN: Extends detection to the exact pixels of each object.
  • Pose Estimation: Locates body points rather than whole objects.

Back to Computer vision · All topics

The interactive demonstration and the Python walkthrough for this topic need JavaScript.