Pooling is a downsampling technique in CNNs that reduces spatial dimensions while retaining key features. Common pooling types include max pooling and average pooling. Pooling helps control overfitting by providing a form of translational invariance.