Convolutional Neural Networks

Researchers came up with the concept of CNN or Convolutional Neural Network while working on image processing algorithms. Traditional fully connected networks were kind of a black box — that took in all of the inputs and passed through each value to a dense network that followed into a one hot output. That seemed to work with small set of inputs.

But, when we work on a image of 1024x768 pixels, we have an input of 3x1024x768 = 2359296 numbers (RGB values per pixel). A dense multi layer neural network that consumes an input vector of 2359296 numbers would have at least 2359296…