AI engineering glossary · Computer vision
What is cross-correlation?
Cross-correlation: The same sliding-window sum as convolution but without flipping the kernel first. Deep learning libraries actually implement this operation and call it convolution; since kernels are learned, the flip makes no practical difference.
Cross-correlation: an example
PyTorch's Conv2d computes cross-correlation, so a hand-designed edge kernel copied from a signal-processing textbook may produce a mirrored result.
Related terms
More Computer vision terms
- CutMix
- Cutout
- Data augmentation for images
- DeepLab
- Depthwise separable convolution
- Dice coefficient
- Dilated convolution
- Document layout analysis
Where you learn it
Cross-correlation is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.