(NOTES) Convolutional Neural Networks

4. Convolutional Neural Networks:

Slides

4.1 Week 1 - Convolutional Neural Networks

4.1.1 Computer Vision

4.1.2 Edge Detection Example

4.1.3 More Edge Detection

4.1.4 Padding

4.1.5 Strided Convolutions

4.1.6 Convolutions Over Volume

4.1.7 One Layer of a Convolutional Network

4.1.8 Simple Convolutional Network Example

4.1.9 Pooling Layers

4.1.10 CNN Example

4.1.11 Why Convolutions?

4.2 Week 2 - Case Studies / Practical Advice for Using ConvNets

4.2.1 Why look at case studies?

4.2.2 Classic Networks

4.2.3 ResNets

4.2.4 Why ResNets Work?

4.2.5 Networks in Networks and 1x1 Convolutions

4.2.6 Inception Network Motivation

4.2.7 Inception Network

4.2.8 MobileNet

4.2.9 MobileNet Architecture

4.2.10 EfficientNet

4.2.11 Using Open-Source Implementation

4.2.12 Transfer Learning

4.2.13 Data Augmentation

4.2.14 State of Computer Vision

4.3 Week 3 - Detection Algorithms

4.3.1 Object Localization

4.3.2 Landmark Detection

4.3.3 Object Detection

4.3.4 Convolutional Implementation of Sliding Windows

4.3.5 Bounding Box Predictions

4.3.6 Intersection Over Union

4.3.7 Non-max Suppressionk

4.3.8 Anchor Boxes

4.3.9 YOLO Algorithm

4.3.10 Region Proposals (Optional)

4.3.11 Semantic Segmentation with U-Net

4.3.12 Transpose Convolutions

4.3.13 U-Net Architecture Intuition

4.3.14 U-Net Architecture

4.4 Week 4 - Face Detection / Neural Style Transfer

4.4.1 What is Face Recognition?

4.4.2 One Shot Learning

4.4.3 Siamese Network

4.4.4 Triplet Loss

4.4.5 Face Verification and Binary Classification

4.4.6 What is Neural Style Transfer?

4.4.7 What are deep ConvNets learning?

4.4.8 Cost Function

4.4.9 Content Cost Function

4.4.10 Style Cost Function

4.4.11 1D and 3D Generalizations