Object detection is the computer vision task of finding each object in an image, drawing a box around it and naming what it is.
In plain words
Classifying an image tells you what is in it; detection also tells you where. A detector outputs a list of boxes, each with a label and a confidence score. Models such as YOLO do this in a single pass over the image, fast enough to run in real time. Detection is useful when you need location but not exact shape.
A quick example
A shop camera feeds a detector that outputs boxes labelled person, trolley and shelf, each with a score like 0.93. Software counts the person boxes to measure queue length at the tills.
Where you'll meet it
- Real-time detection in camera and video feeds.
- Counting and locating objects in photos.
Related
Sources
- PaperYou Only Look Once: Unified, Real-Time Object Detection, Redmon et al. (arXiv) · read 29 Sept 2026
- DocsObject Detection, Ultralytics documentation · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 31 Oct 2026. Spotted an error? Report it.