Skip to content
BENIBIX Learn

You Only Look Once (YOLO)

Also known as YOLO, YOLO Model, YOLO Object Detection

A fast model that finds objects in a photo. It finds them all in one look.

Think of it like

It is like a quick glance at a room. One look, and you know where the chairs and people are.

Example

A shop camera runs a small YOLO model. It boxes each shopper in every frame of live video, so staff can see how busy the aisles are.

Why it matters

Robots and cars must see fast. YOLO made it possible to find objects in live video on cheap computers.

Definition

YOLO is a family of object detection models that look at a picture just once. Older ways looked at many small parts of an image, one after another. YOLO splits the image into a grid and finds every box and label in one pass. That makes it fast enough for live video. The first paper came out in 2015, and many newer versions have come since.

Example code

bash
pip install ultralytics
yolo predict model=yolo26n.pt source='https://ultralytics.com/images/bus.jpg'
Two lines install a YOLO tool. Then it boxes the people and the bus in a test photo.
On the left, YOLO lays a grid on the photo. It finds every box in one pass.

Common mistake

Thinking YOLO is one model. It is a long line of versions from different teams, and they differ a lot.

Key takeaways

  1. 1
    One pass. The whole image is read at once.
  2. 2
    Grid of guesses. Each cell predicts nearby boxes.
  3. 3
    Built for speed. Fast enough for live video.