AI engineering glossary · Computer vision
What is roI pooling?
RoI pooling (also called RoI Align): A layer that takes a region of any size from a feature map and converts it to a fixed-size grid, such as 7 x 7, by max pooling in sub-cells. It lets a fixed classifier handle variable-size proposals; RoI Align is a more precise variant.
RoI pooling: an example
A 23 x 41 region is divided into a 7 x 7 grid of cells, and each cell's maximum forms the output.
Related terms
More Computer vision terms
- Semantic segmentation
- SIFT
- Single-stage detector
- Sobel operator
- Stride
- Text detection
- Text recognition
- Thresholding
Where you learn it
RoI pooling is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.