setpixel

code from class 11, 12

here’s code from class 11 (various opencv advanced examples) and class 12 (using warpIntoMe and ArToolkit / fiducial tracking):

class 11 /  class 12

happy hacking!

wanna comment?

code from class 10

wanna comment?

homework 7

a) Build an application that tracks multiple colors (instead of just 1).  Use the color tracker code and extend it to track two or more colors.  You may need additional grayscale images, per color you are tracking.   Think about how you might build a “general purpose” color tracking class.   Also, you might need multiple panels per color (you can add panels in the setup of the control panel).

b)  Use the face tracker to make some sort of interaction that is controlled via face position and/or size.  For example, can you make animations that change based on your proximity to the camera?

c) As an inspiration for your 2nd big project, find 5 examples of “unique / alternative” drawing interfaces that use technology in interesting ways.  For example, drawdio  withDRAWl  etc.   There doesn’t need to be a computer vision element in play in these projects, but it’s good if they are pushing the idea of drawing and interface.

wanna comment?

code from class 9

color / text / face : class9

(note — I’ve update the color tracker code since there was a bug in 0.06… should be all good now)

wanna comment?

homework 6

So, first of all, GREAT JOB with the trial by fire.  Now, within your new groups, I’d like you to make two computer vision based projects.   Please check out

http://muonics.net/cvMovies

for test footage.

you’ll be making two different projects:

a) The first is an update on the “daisies” application that was made by  Theo in my first setFrame() computer vision class. http://www.theowatson.com/site_docs/work.php?id=18
Please think about what a 2009 version of this kind of application might look like.

b) make a comptuer vision application that uses a very specific kind of action – such as moving in a circle (as in the spinning vinyl application) or this brainfu*k body interpreter:
http://vimeo.com/7133810

please push yourself and your groups to make expressive, engaging applications.

wanna comment?

parts list for computer vision (ps3)

this is to make an IR ps3 camera –

• Ps3 Camera:
  http://is.gd/4K8dR
  (you can find this at best buy, toys-r-us, etc)
• Mounts (after you take apart the camera as described here)Perfect (but expensive)
  http://peauproductions.com/store/Almost Perfect (cheap)
  http://sunex.com/ (type is CMT107_SU Lens holder, M12×0.5, 15.5, centered)
  these required a bit of scraping to make the holes wide enough for the screws.
  
• Lens Sets:
  dealextreme.com
  http://www.dealextreme.com/details.dx/sku.15774
  (Item# 15774 .8mm~16mm Fixed IRIS Lens Set for Webcams and Security/CCTV
  Cameras (6-Lens Pack)

• IR
  IR filter here (same as kodak wratten, but way cheaper):
  freestylephoto.biz
  5331007 LEE IR 87C 100X100MM FILTER INFRARED 4X4
  $26.9900
  

then, on a mac, install MACAM, then you need to swap in the modified maccam component (http://www.openframeworks.cc/files/macam-ps3eye-manualwb-defaultVGA.zip) to /Library/QuickTime/   (/Library is at the root of your hard drive, not your user directory), restart the machine, to get a good, fixed everything ps3 camera image from their computer.

for windows, you’ll need to install the direct show driver:
http://createdigitalmotion.com/2009/08/27/ps3-eye-camera-drivers-updated-for-windows-fixes-performance-options-awesomeness/

wanna comment?

code from class 8 (opencv presence and motion)

here is presence and motion detection code from class

class8

wanna comment?

code from class 7 (opencv intro)

this is the intro to openCV code we covered in class.  Please have fun experimenting with it:   class7

wanna comment?

connected components / contour extraction crash course

here’s the set of images that I’d like your groups to process.  For each one, do:

a) erosion
b) dilation
c) connected components (display the components somehow, and also, report back how many there are)
d) contour extraction (draw the contour as a set of lines)

wanna comment?

homework 4

two things:

a) Create a poster based on interpreting an image.  think about things like the border of perception, the cultural significance of images (such as we’ve seen in the work of Jason Salavon and Nancy Burson) and how you can crate unique algorithms to express yourself via processing an image.  you can use the postscript output (or diffrerent means) but your poster shouldn’t be a singl piece of 8.5×11 paper.  Come to class carrying something unique that you are proud of.

b) Median filtering.  Given the pixel + neighbors routine we talked about in class (including the edge conditions and all of the n/e/s/w type code, do the following:

for every pixel in an image, put it and it’s 8 neighbors (using the edge fix we discussed in class) in a temporary array of 9 elements, that you sort with qsort.  then take the middle most element (the median) and use that as the new value.

try doing both gaussian blur (1,2,1 / 2,4,2 / 1,2,1 kernel) and media filtering on a noisy image, at least once but perhaps multiple times.  Which is better at fixing or removing points of noise?

wanna comment?