From FutureTheater Wiki
Some of the material here is taken from my article, Computer Vision for Artists and Designers.
Here is a compilation of notable interactive artworks that use computer vision.
- Myron Krueger, Videoplace (1974)
- Camille Utterback & Romy Achituv, Text Rain (1999)
- Chunky Move, Mortal Engine (2006)
- Philip Worthington, Shadow Monsters (2006)
- Tmema (Levin + Lieberman), Messa di Voce (2003)
- Klaus Obermaier, Apparition (2005) (Also)
- Art+Com, Jew of Malta (2002)
- David Rokeby, Sorting Demon
- Christian Moeller, Cheese (2003)
Various works by Scott Snibbe:
Various works by Rafael Lozano-Hemmer:
- Subtitled Public (2005)
- Underscan (2005)
- Standards and Double-Standards (2004)
- Surface Tension (1993)
Here is the bundle of downloads we will use for today's workshop.
Simple image processing
- Accessing pixels
- Simple thresholding (File/Examples.../Libraries/Video/BrightnessThresholding)
- Simple background subtraction (In download bundle)
- Background subtraction (File/Examples.../Libraries/Video/BackgroundSubtraction)
- Frame differencing (File/Examples.../Libraries/Video/FrameDifferencing)
- Brightness tracking (File/Examples.../Libraries/Video/BrightnessTracking)
And then some applications:
- LaserTag (In download bundle)
- TextRain (In download bundle)
- SlitScanTiny (In download bundle)
Higher-level analyses using OpenCV
- Blob segmentation & Contour extraction (In download bundle)
- Face detection (In download bundle)
For the OpenCV examples, we'll use the OpenCV library for Processing.
Complete instructions are there. I have already included You'll need to download three things from that site:
- (For MacOSX) Download and install the opencv-framework-1.1.dmg
- Download, unzip, and move the opencv_01.zip (OpenCV Library) into your Processing libraries folder.
- Download and unzip the opencv_examples.zip.
Computer vision in Processing with the Kinect
Further possibilities for computer vision are made possible with the Kinect, a "depth camera" that can sense how far away each pixel is from the camera, in addition to its RGB color. Daniel Shiffman from NYU has created a Processing interface for the Kinect; his complete instructions are here: http://www.shiffman.net/p5/kinect/
- Download the following: https://github.com/shiffman/libfreenect/raw/master/wrappers/java/processing/distribution/openkinect.zip (These materials are also included in the Workshop Bundle for your convenience.)
- With your Processing app closed (not running), unzip and install the openkinect folder into your Processing "libraries" folder, i.e. Sketchbook/libraries/openkinect
- Now launch Processing. Check that the library is installed properly by going to: Processing->Sketch->Import Library->openkinect
- You can now open and run Daniel's examples: Processing->Open...
There are too many great resources out there to count, but I happen to find the following sites to be quite useful.
- Fisher, Robert, et. al. HIPR (The Hypermedia Image Processing Reference).
- Fisher, Robert, et. al. CVonline: The Evolving, Distributed, Non-Proprietary, On-Line Compendium of Computer Vision.
- Huber, Daniel et. al. The Computer Vision Homepage.