CNNs cemented the position as the de facto model for computer vision with the introduction of VGGNet, ResNe(X)t, MobileNet, EfficientNet, and RegNet.