Search⌘ K
AI Features
Home
Courses
Deep Learning for Android Apps

Convolutional Neural Networks

Learn how convolutional neural networks work, including key concepts like convolution operations, feature maps, and CNN layers such as convolutional, pooling, normalization, and fully connected layers. This lesson helps you understand CNN architectures to deploy effective deep learning models on Android devices using TensorFlow Lite.

We'll cover the following...

What is a convolution?

Convolution is a linear mathematical operation that combines two signals or mathematical functions, f(x)f(x) and g(x)g(x), to produce an output signal, h(x)h(x). Mathematically:

The symbol ∗* denotes a convolution operation, whereas the rightmost expression involving a summation is a convolution sum. Convolution is usually employed to find the output response h(x)h(x) against the input signal f(x)f(x) when passed through a kernel or a filter g(x)g(x).

Two-dimensional convolution

Digital images are two–dimensional signals that have two spatial independent variables: xx ...