Lesson List
Section 1. Course Intro – Activity 1: FaceBoxes – Basic Face Detection
In this brief section, including Activity 1 we will discuss the various topics we will cover in this course, as well as some background on the concept of artificial intelligence vs natural intelligence.
0/4
Section 2. History of Computing & AI – Activity 2: What is AI?
In this section, we will discuss the early and recent history in AI technology development. We will also cover some background on semiconductors, CPU's, GPU's and other technology that make AI possible. We will also discuss the basics of AI model training, and go through some of today's most popular AI technologies.
0/1
Section 3. How is AI used in different applications? – Activity 3: Aligning real world AI with AI4K12
In this section, we will discuss several real world examples of AI robotics, and go through the exercise of aligning AI concepts with the AI4K12 5 Big Ideas in AI.
0/1
Section 4. AI in action: Teachable Machine – Activity 4: Training and testing an AI model
In this section, we will conduct a browser based activity, and will go through all the steps of developing training data, uploading it to a Google server, and training a real AI model, for object detection. We will then test our resulting model with different objects and record our results, to determine if the model is well-built and if there are any potential drawbacks or ways to improve the system.
0/2
This feature has been disabled by the administrator
Lesson: FaceBoxes: Simple face detection demonstration & activity
Overview
Exercise Files
About Lesson

Activity 1: FaceBoxes

Brief Face Detection Activity

  1. Make sure your webcam is attached and that no other application is using it. If so, turn that application off.
  2. Draw several face pictures, on a piece of scrap paper, using markers or pens (something with thick lines is preferable). Any style is fine, be creative and try different things.
  3. Visit the CodePen OpenCV Face Detection Program: https://codepen.io/bmonteith/full/BaJWzXx
  4. Show your artistic faces to the webcam, and check out the response.
  5. Record your observations in a chart, similar to the one below.

 

Face

Did it work?

(meaning, did a box show up?)

How well did it work (or other observations)

yes

Works OK but not as good as the bone below; maybe the marker isn’t dark enough

yes

best one, had the best response, could be the marker color

no

Maybe because it has no hair, and it’s at an angle, not straight to the camera

yes

Maybe it works because it has hair, but even though it’s at an angle it works even though the one above doesn’t

 Things to think about:

  • Which faces worked and which ones didn’t work?
  • Why do you think that is?
  • If it can’t recognize every type of face, is there a problem with the data that was used to make this AI?
  • How could this program be improved?
Exercise Files
No Attachment Found
No Attachment Found
Lesson List
Section 1. Course Intro – Activity 1: FaceBoxes – Basic Face Detection
In this brief section, including Activity 1 we will discuss the various topics we will cover in this course, as well as some background on the concept of artificial intelligence vs natural intelligence.
0/4
Section 2. History of Computing & AI – Activity 2: What is AI?
In this section, we will discuss the early and recent history in AI technology development. We will also cover some background on semiconductors, CPU's, GPU's and other technology that make AI possible. We will also discuss the basics of AI model training, and go through some of today's most popular AI technologies.
0/1
Section 3. How is AI used in different applications? – Activity 3: Aligning real world AI with AI4K12
In this section, we will discuss several real world examples of AI robotics, and go through the exercise of aligning AI concepts with the AI4K12 5 Big Ideas in AI.
0/1
Section 4. AI in action: Teachable Machine – Activity 4: Training and testing an AI model
In this section, we will conduct a browser based activity, and will go through all the steps of developing training data, uploading it to a Google server, and training a real AI model, for object detection. We will then test our resulting model with different objects and record our results, to determine if the model is well-built and if there are any potential drawbacks or ways to improve the system.
0/2
This feature has been disabled by the administrator
0% Complete