Covers: theory of Convolutional Neural Network
Estimated time needed to finish: 20 minutes
Questions this item addresses:
  • How can I apply Sobel and Scharr Operator on image?
  • What is the difference between Sobel and Scharr Operator and how can I visually compare them?
  • What is a Convolutional Neural Network (CNN)?
  • How can one understand a CNN?
How to use this item?

In this video you will learn about:

  • Sobel and Scharr operators application on image ConvNets and their applications
  • Convolutional Neural Network (CNN)

What is a Convolutional Neural Network (CNN) and how to understand it?

A CNN is a type of NN where we apply convolutional layers to our net usually to extract visual features. We can make a CNN architecture in many ways, they are similar to ordinary NN because they are made up of learnable weights (convolution kernels) and biases. Alt text

Now, look at the specific one layer. Notice the sliding kernel. Also notice that, even though it's not obvious on first sight, the dimensionality of the projected matrix (matrix on the right) is lower then the matrix (the input matrix on the left) upon which the kernel is sliding on. You can observe that that the edges of the later "A" are closer to the edges on the projected matrix than they are on the input matrix. slidingkernel

In 2012, Alex Krizhevsky in collaboration with Ilya Sutskever and Geoffrey Hinton developed a CNN called AlexNet. It was a revolution in CV. The model became an inspiration for future state-od-the-art works! Alt text

Fail to play? Open the link directly: https://youtu.be/azR3pOV5K8I
Author(s) / creator(s) / reference(s)
Amir Hajian
0 comment
Recipe
publicShare
Star(0)

Extracting Features Using Convolution And Regularization

Contributors
Total time needed: ~4 hours
Objectives
Feature extraction via convolution kernels
Potential Use Cases
Building a DL architecture for for computer vision model.
Who is This For ?
INTERMEDIATE
Click on each of the following annotated items to see details.
Resources6/7
VIDEO 1. Mathematics of Convolution
  • Why wee need more than MLP?
  • What is a convolution and kernel?
  • What is a Hann function and how to apply it?
  • What are ConvNets and what is their architecture?
31 minutes
VIDEO 2. Why Convolutions? Sobel & Scharr Filters
  • Why to use Convolutions?
  • What is an image filter?
  • What is image segmentation and how is it done?
  • What convolution does really do?
  • What are two famous and edge detection filters / algorithms?
16 minutes
VIDEO 3. 2D Convolutions, Pooling, and Dilated Convolutions
  • What is pooling and padding?
  • What other 2D convolution techniques do exist?
  • When do we use padding and how?
  • When do we use pooling and how?
  • What are dilated convolutions and how they differ from standard convolutions?
30 minutes
VIDEO 4. Conv-Nets
  • How can I apply Sobel and Scharr Operator on image?
  • What is the difference between Sobel and Scharr Operator and how can I visually compare them?
  • What is a Convolutional Neural Network (CNN)?
  • How can one understand a CNN?
20 minutes
VIDEO 5. Regularization using Dropouts
  • Why do we need dropouts?
17 minutes
REPO 6. Hands-on Convolutional Networks
  • How to implement convnets in PyTorch?
30 minutes
RECIPE 7. Understanding Convolution
40 minutes

Concepts Covered

0 comment