AI engineering glossary · Computer vision
What is anchor box?
Anchor box (also called Prior box, Default box): A predefined reference box of a chosen size and aspect ratio placed at many positions on the image. Detectors predict, for each anchor, whether it holds an object and how to adjust it, which simplifies handling many object shapes.
Anchor box: an example
At each feature-map cell, a detector might use 9 anchors: three scales times three aspect ratios.
Related terms
- Bounding box regression
- Region proposal network
- Aspect ratio
More Computer vision terms
- Aspect ratio
- Average pooling
- Binary mask
- Bottleneck block
- Bounding box
- Canny edge detector
- Channel
- CIFAR-10
Where you learn it
Anchor box is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.