Object Detection
Last Updated: July 29, 2026 | By Mihail Sebastian | AI Dictionary
A computer vision task that finds every object in an image and marks where each one is, returning a class label and a bounding box for each detection.
What is Object Detection?
Object detection is a computer vision task that finds the objects in an image and marks where each one is, returning a class label and a bounding box for every object it detects. It answers two questions at once: what is in this image, and where.
The “where” is the point. Systems that act on the physical world – a car braking for a pedestrian, a robot picking a part from a bin – need locations, not just labels.
How Object Detection Works
Detectors come in two families. Two-stage models such as Faster R-CNN first propose regions likely to contain objects, then classify each region; they are accurate but relatively slow.
Single-pass models such as YOLO and SSD predict all boxes and labels in one run through the network, trading a little accuracy for real-time speed. Both families finish with non-max suppression, a step that removes duplicate boxes covering the same object.
Object Detection vs Image Recognition
Object detection labels each object and marks its position with a bounding box; image recognition labels the whole image and stops there. Shown the same street scene, a detector returns three cars, two pedestrians, and a traffic light, each with coordinates; a recognition model returns “street.”
| Criterion | Object detection | Image recognition |
|---|---|---|
| Question answered | What is in the image, and where? | What does the image show? |
| Output | A class label plus a bounding box per object | Label(s) for the whole image |
| Multiple objects | Each object found and located separately | One set of labels, no locations |
| Typical use | Driver assistance, video analytics | Photo tagging, content moderation |
Example of Object Detection
A city traffic system runs a detector on every camera frame. For one frame, the model returns a list: a car at 98% confidence with its box, a second car at 95%, a pedestrian at 91%.
Downstream software does the rest. It counts vehicle boxes per lane to time the signals, and it raises an alert when a pedestrian’s box overlaps the roadway outside a crossing. The detector supplies labeled coordinates; the logic that acts on them is ordinary code.
Related AI terms: YOLO · Image Recognition · Computer Vision · Convolutional Neural Network
Did you like the Object Detection gist?
Learn about 250+ need-to-know artificial intelligence terms in the AI Dictionary.
Mihail Sebastian — Writes about AI governance, regulation, and the technology behind them. Placeholder bio — replace with a real credential line. About