The Parallax View

Joel Gethin Lewis

Lecture 3: Computer Vision 1

What I'm going to talk about today:

1. Thresholding images into different parts.
2. Pixel differences between frames of video.
3. Live tracking of colours, markers and faces.
4. Brief: letting your computer see.

1. Thresholding images into different parts.

• Today we are going to be exploring the field of Computer Vision. Binary Images are used widely in Computer Vision to select areas of interest. One way of producing a Binary Image is to Threshold a Greyscale image. Image segmentation is the technical term for breaking up an image into different areas of meaning.
• Throughout this lecture we'll be drawing on Kyle McDonald's cv-examples. In session 5, we'll be connecting with him live.
• Average Brightness.
• Thresholding.
• Brightest Point.

2. Pixel differences between frames of video.

• Moving from analysing single images to the differences between frames allows a variety of new features to be detected - from edges to movement (or flow) within an image.
• Background Subtraction.
• Frame Difference.
• Motion History.
• Point Tracking.
• Optical Flow.
• Edge Detection.
• Contour Detection (trackingjs).

3. Live tracking of colours, markers and faces.

• Doing even more complex analysis allows us to track colours, markers and even faces.
• Colour Detection.
• Marker Tracking.
• Face Detection.
• Face Tracking.

4. Brief: letting your computer see.

• In your pairs, what do you want your computer to see? Why? When? Where? For whom?
• You live in a visual culture, what do you want to question about it?
• How to See the World by Nicholas Mirzoeff - a book on visual culture.
• Feedback is when you feed the output of a signal into the input of it. It's a very powerful technique that can be used in a variety of fields:
• Video feedback, the video equivalent.
• Nam June Paik, TV-Buddha. Media Art Net is a great resource.
• Douglas Trumbull, 2001 Star Gate Sequence.

4. Brief: letting your computer see (continued).

• An Informal Catalogue of Slit-Scan Video Artworks and Research by Golan Levin. Golan's portfolio. Slit-scan demo in p5js by Daniel Shiffman.
• Going around the corner piece by Bruce Nauman.
• Tell the story of making audio reactive visuals for Massive Attack in 2008.
• Finally, a great Twitter thread on the history of new media art from Greg J. Smith.
• All works must be published on the brand new p5.js Web Editor.

Thanks!

• More about me: @joelgethinlewis
• Get in contact via: www.joelgethinlewis.com
• Website:
  https://jgl.github.io/TheParallaxView/
• Today's slides:
  https://jgl.github.io/TheParallaxView/lecture3.html