Introduction to Convolutional Neural Networks - Part 1 - Lecture 6

Dive into the fundamentals of Convolutional Neural Networks (CNNs) in this comprehensive lecture from the University of Central Florida's CAP5415 course. Begin with an introduction to CNNs, exploring their history, significant breakthroughs, and widespread applications beyond image processing. Examine the general CNN architecture, revisiting key concepts such as filtering, correlation, and convolution. Compare traditional neural networks with CNNs, and delve into the intricacies of convolution layers, operations, and parameters. Refresh your understanding of edge detection techniques and witness practical demonstrations. Gain insights into 2D convolution dimensions, stride, padding, and pooling. Conclude by learning how to visualize CNNs, providing a solid foundation for further exploration of this powerful deep learning technique.

Intro
A quote from a famous scientist...
CNN - example: depth estimation
Convolutional Neural Network (CNN)
History
First Strong Results
Today: CNNs are everywhere
CNN - Not just images
General CNN architecture
Filtering - recap
Correlation (linear relationship) - recap
Convolution -recap
Learning phases
Neural Network vs CNN
Convolution layer
Convolutional Network
Parameters
Convolution Operation
Sobel Edge Detector -recap
Demo
Convolution - Intuition
2D Convolution - dimensions
Stride
Padding
Pooling
Visualizing CNN