AI Glossary

Convolutional neural network

Also called CNN, ConvNet

What does “Convolutional neural network” mean?

Convolutional neural network: Built for grid-like data such as images, a CNN slides small learned filters across the input to detect local patterns like edges and textures, then combines them into higher-level features. It dominated computer vision for roughly a decade before transformers became common.

Convolutional neural network visual guide: Built for grid-like data such as images, a CNN slides small learned filters across the input to detect local patterns like edges and textures, then combines them into higher-level features. It dominated computer vision for roughly a decade before transformers became common.
Visual guide Open full size (new tab)

What is an example of “Convolutional neural network”?

ResNet is a widely used CNN architecture, and CNNs still run many on-device vision tasks because they are efficient.

Keep learning

See also

Vision transformerTermRecurrent neural networkTermComputer visionTerm

Same topic

Style transferTermDecoderTermBERTTermDense modelTermAutoregressive modelTerm