Day 7: Computer Vision

Recap

  • Learned about neural networks and how they work
  • Used Face Detection, Object Detection, and Pose Estimation CV algorithms
  • Mapped Pose Landmarks to a 3D scene

Two More CV Algorithms

Gesture Recognition

  • Question: “Where gesture am I making with my hands?”
  • Detects key landmarks on a hand — wrist, palm, fingers
  • Categorizes into one of 8 different gestures: closed fist, open palm, victory, etc.
  • Examples:
    • Game or device control with hands
    • Sign language detection
    • Virtual art applications

Demo

Gesture Recognition Notebook

Hands-On

Gesture Recognition Notebook

Demo

Gesture Painter Notebook

Hands-On

Gesture Painter Notebook

Demo

Rock, Paper, Scissors Notebook

Hands-On

Rock, Paper, Scissors Notebook

Image Segmentation

Image Segmentation

  • Question: “Which pixel belongs to which object?”
  • Goes beyond bounding boxes — labels every single pixel
  • Examples:
    • Photo apps removing backgrounds with one tap
    • Virtual wardrobe or hair/make-up
    • Autonomous vehicles painting a full picture of the road scene

Demo

Image Segmentation Notebook

Hands-On

Image Segmentation Notebook

Create Your Own

Create Your Own

  • For the rest of today, create your own mini-app
  • Pick any of the 5 CV algorithms
  • Copy/paste the examples
  • Add more Python code to make it your own
  • Ask for help if you are stuck!

Tomorrow

  • Introduce Generative AI
  • Customize a chat bot of our own
  • Learn how chatbots can create music and 3D!