Hi. I've got this fun-looking background difference thing going on (code below). It looks cool, but isn't quite what I want, which is something more like a vertex-based shape that I can then alter and mess with as I choose. I know reading through the single-dim pixels array, and running a brightness test on that pixel, turning it into a vertex if it passes, etc. doesn't do it (obviously a for() loop won't read the array in the correct direction for this to work properly). So, anyone have any ideas? I'm trying to take the moving shape it's detecting and be able to control it as a shape.

Code:

import hypermedia.video.*;
OpenCV opencv;
int numPixels;
int blackval;

void setup() {
  size(320,240);
  opencv=new OpenCV(this);
  opencv.capture(width,height);
  numPixels=opencv.width*opencv.height;
  colorMode(HSB,100);
}

void draw() {
  background(0);
  opencv.read();
  points();
}
  
void points() {
  opencv.absDiff();
  opencv.blur(OpenCV.BLUR,5);
  
  int[] cap=opencv.pixels();
  loadPixels();
  
  for(int i=0; i<numPixels; i++) {
    if(brightness(cap[i])>7) {
	cap[i]=color(0,0,100);
	
    }
  }
  updatePixels();
  
 PImage newcap = createImage(width,height,HSB);
   newcap.loadPixels();
   for (int i = 0; i < numPixels; i++) {
     newcap.pixels[i] = cap[i];
   }
   newcap.updatePixels();
   image(newcap,0,0);
  
  filter(INVERT);
  opencv.remember();
}