AI Blog

AI Blog

by Michele Laurelli

Filter / Kernel

/ˈfɪltər ˈkɜːrnəl/
Concept
Definition

A small matrix of learnable weights that slides over input during convolution to detect specific features.

Filters are learned during training to detect useful features. Common sizes: 3x3, 5x5, 7x7. Multiple filters extract different features. The term 'kernel' is used interchangeably with filter.

Examples

1

3x3 edge detection filter

2

5x5 Gaussian blur kernel

3

Learned feature detectors

Michele Laurelli - AI Research & Engineering