Vision

What is YOLO?

YOLO is a family of real-time object detectors, including YOLOX, that predict boxes and classes on stills or video frames.

Updated · Part of What is object detection?

How it works

  • Run it on an image or a frame row.
  • Read boxes, classes, and scores.
  • Filter frames by the classes you care about.

What it is not

It is not CLIP retrieval, and it is not a segmentation mask model.

YOLO: this, and the thing it is confused with

YOLO: this, and the thing it is confused with
ThisNot this
SpeedBuilt for real timeA two-stage detector’s default
OutputBoxesA mask or a caption
On videoOnce per sampled frameOnce on the whole file

Where Pixeltable fits

Pixeltable calls YOLOX as a function on the frame column, for example yolox(frame, model_id=...).

Questions

How does YOLO work?
Run it on an image or a frame row. Read boxes, classes, and scores. Filter frames by the classes you care about.
What is YOLO often confused with?
It is not CLIP retrieval, and it is not a segmentation mask model.