EV Graduate Seminar - AD508 F08

My initial intentions for my final project was to achieve a real time conversion of video into what would look somewhat like a Kirlian photograph.  Kirlian Photography is accomplished by electrically charging the film  as it is exposed.  It comes in two forms, one which looks like lightning along the exposed edges of a sillouette, the other looks like hazy colored fog.  I decided to go for the lightning aesthetic.  To that goal, I did get a working prototype.  It generated pixel positions based off of differences in contrast in light and drew lines a curves some what perpendicular to the normal of those pixel positions.  This created lightning like edges.

Though technically, my prototype was successful, I did not really achieve the aural energy body feel that I was going for. As I plan on using this a starting point for a supporting piece for my thesis show, I have been scraping my brain how to bring this to the next level.  I plan on reimplimenting this script using IR cameras instead of visible light.  I feel this is the next important step to push the etheral quality of the energy body aesthetic I am tring to create.  At the very least, the transformation of unseen to seen will bring the project into a closer connection to simulated aural visualization.

Had a little trouble finding it but here’s the code zipped up.

http://www.gravatonia.com/assignment01_grav.zip

tkosmalski_feathers

vertex_20

Wiimote-Based Video Control System Video Documentation
Adam Trowbridge

Test of Wiimote-Based Video Control System. The system is written in Max/MSP/Jitter and takes advantage of Andreas Schlegel’s OSC version of Hiroaki Kimura’s darwiinRemote.

http://code.google.com/p/darwiinosc/downloads/list
http://sourceforge.net/projects/darwiin-remote/

The system takes advantage of the Wiimote’s orientation data, transmitted to a laptop via bluetooth. The data is received in Darwiin Wiimote and transmitted to Max/MSP/Jitter via OSC (Open Sound Control). In Max, the Y-axis orientation data is parsed and matched to the frames of the video so that turning the Wiimote all the way to one side moves the playhead to the beginning and, conversely, turning it to the other side moves the playhead to the end. It is thus possible to shake or rock the Wiimote and achieve a forward/reverse loop point.

www.atrowbri.com

I’m attaching my final project documentation to this post.

project-document-ad508–arunan-rabindran

Arunan Rabindran

Hidden Horizon

Description:

In today’s urban landscape the horizon cannot be seen, it is always hidden by buildings, houses, advertisement, etc. There is almost no point in the city where the human sight can go beyond a hundred meters in horizontal direction.

A series of panoramic windscreen projections will be assembled to cover the width of a building. Live video feed will be transformed into horizontal lines of colors with an algorithm (an adaptation of Ben Fry’s Disgrand) . The position of the generated horizon will be the same of the hidden horizon.

overview-of-sage

The link above should allow download of my paper

“Overview of SAGE (Scalable Adaptive Graphics Environment) for Artists, and Critique of its Associated Technical and Other Documentation”

Andrew Oleksiuk

I am including the code and links in this post.    The philosophical stuff will come this evening.  The first link is a small compact video.  Quality is not fabulous, (IMHO) but it communicates the process.  The second and third video links are much larger (1080p) and raw footage, but you can make out the details a lot better.

http://www.gravatonia.com/Weldon_final.mov

http://www.gravatonia.com/raw1.MOV

http://www.gravatonia.com/raw2.MOV

And here is the final (for now) version of the code:

import JMyron.*;

JMyron m;

void setup(){
int w = 640;
int h = 480;
frameRate=15;
size(w,h);
m = new JMyron();
m.start(640,480);
m.findGlobs(1);
println(”Myron ” + m.version());
}

void mousePressed(){
m.settings();
}

void draw(){
m.trackColor(255,255,255,255);

m.update();
int[] img = m.image();

//first draw the camera view onto the screen
loadPixels();
for(int i=0;i<width*height;i++){
pixels[i] = img[i];
}
updatePixels();
background(int(random(5)),int(random(5)),int(random(5)));
//draw edge pixels of globs (this and the next chunks of code are chokers)
int list[][][] = m.globPixels();
stroke(235+int(random(15)),240+int(random(10)),250);
for(int i=0;i<list.length;i++){
int[][] pixellist = list[i];
if(pixellist!=null){
beginShape(POINTS);
for(int j=0;j<pixellist.length;j++){
if(j%2!=0){
line( pixellist[j][0]+int(random(4)),  pixellist[j][1]+int(random(4)),pixellist[j][0]+int(random(-4)),  pixellist[j][1]+int(random(-4)));
}
if(j%5==0){
line( pixellist[j][0]+int(random(6)),  pixellist[j][1]+int(random(6)),pixellist[j][0]+int(random(-6)),  pixellist[j][1]+int(random(-6)));
}
}
endShape();
}
}

}

public void stop(){
m.stop();
super.stop();
}

Though it does run at 24 and 30 FPS, I found it smoothest at 15 and 18 FPS.  Only weakness is it tended to fall apart if I went over 720 by 480 and I could figure out how to get it to go full screen.  Suggestion on that one would be HIGHLY appreciated.

Grav