From the course: AI Workshop: Hands-on with GANs using Deep Convolution Networks

Unlock this course with a free trial

Join today to access over 25,500 courses taught by industry experts.

Understanding convolutional neural networks

Understanding convolutional neural networks

From the course: AI Workshop: Hands-on with GANs using Deep Convolution Networks

Start my 1-month free trial Buy for my team

Understanding convolutional neural networks

“

- [Instructor] Let's get a big-picture understanding of how convolutional neural networks, or CNNs, work. CNNs are a class of deep learning models primarily used for computer vision tasks. They're designed to automatically and adaptively learn hierarchical features from input images, making them well-suited to tasks such as image classification and object detection. The CNN architecture, like I mentioned before, is meant to mimic the visual cortex of the brain. Essentially, how humans perceive images. CNNs are able to extract data from images into dimensions. So they also get the spatial information present in an image. Dense neural networks with fully-connected layers can also be made to work with images. Essentially, you'll flatten the image data that you'll feed into a dense neural network. However, flattening the image data does not let the network capture the spatial features of images. The location of an object at the top right or the bottom left of an image, that information is…

Contents