Link Search Menu Expand Document
Robotics101

Lecture 8 - Introduction to Computer Vision with Machine/Deep Learning

  • Monday, 9:00 am – 1:00 pm

Overview

Week 8’s lecture will provide a overview on Machine/Deep learning and common workflows.

  • Deep Learning
    • What is Artificial Intelligence, Machine Learning and Deep Learning, and their relation to each other
    • Advantages of Deep Learning over Machine Learning
    • Neural Networks
      • Overview on what are Neural Networks
      • How do neural networks work
        • Nodes
        • Convolutional Neural Nets (CNNs)
  • Deep Learning in CV
    • How Deep Learning can be used for image recognition
    • Types of image recognition
      • Classification
      • Detection
      • Segmentation
  • Object Detection Workflow
    • What is Object Detection and how it works
      1. Data Collection
        • Methods
        • Considerations
      1. Data Cleaning & Preparation
        • Annotation of Images
        • Image Pre-processing
        • Data Augmentation
        • Train-Test Split
      1. Model Training
        • Model Creation
        • Transfer Learning
        • Training Methods
      1. Model Evaluation
        • Training Metrics
        • Visualization Tools
      1. Model Deployment
        • Exporting the model into different formats

While the concepts covered are non-exhaustive by any means, it should cover most (if not all) that a student will need during the duration of the course. With that in mind, students are still encouraged to explore more advanced concepts to broaden what they are able to do in the their projects.

Further information regarding week 8’s prelab and lab will be provided during the lecture.

Lecture resources