Hi

Im not sure of this one, but try doing this to see the index for the cameras:

String[] devices = Capture.list();
println(devices);

Just put it in the last part of the setup() but before you do any openCV stuff.

hi stephen

i can help with this one - was trying to do exactly the same thing the other day.

Theoretically you should be able to get more than one camera working with openCV, but i think part of the problem with the indexing is that the openCV library only recognises a few types of camera. this is where i got stuck too...

what you have to do is load up the different cameras using the processing video library and then kinda graft those feeds into the openCV buffers.

check this code out and let me know how you get on.

hitting the space bar should record the openCV buffer. it isn't perfect yet; i need to figure out how to record two images to the buffer. ideally i'd like to stitch the two video feeds together at the start...

S

Quote:

So I did get the following code working with two copies of the video object and two copies of the OpenCV object.  Shows two camera images side by side and successfully does face tracking.  (haar files must be in the current dir).  Trouble is, it runs at about 1 fps, compared to about 8 or 10 fps for the pure OpenCV version tracking one camera.  Not sure if my coding is optimized or good.  Thanks for getting me started on this.  Thoughts?

import processing.video.*;  
import hypermedia.video.*;

OpenCV ocv1;
OpenCV ocv2;
Capture myCapture;  
Capture myCapture2;  
String[] captureDevices;  
int threshold = 80;
int midX;

void setup()  
{  
 size(640, 240);  
 midX = width/2;  // place to start drawing image from second cam
 println (Capture.list());  
 captureDevices = Capture.list();  
 
 myCapture = new Capture(this, 320, height, 30);  
 myCapture2 = new Capture(this, 320, height, 30);  
 ocv1 = new OpenCV(this);
 ocv1.allocate(320,height);
 ocv2 = new OpenCV(this);
 ocv2.allocate(320,height);
 ocv1.cascade( OpenCV.CASCADE_FRONTALFACE_ALT );  
 ocv2.cascade( OpenCV.CASCADE_FRONTALFACE_ALT );
 
}  


void draw() {  
 if (myCapture.available()) {  
   myCapture.read();  
   ocv1.copy(myCapture);  // copy to OpenCV buffer
   image(ocv1.image(),0,0); // display to screen
   Rectangle[] faces1 = ocv1.detect();// detect anything ressembling a FRONTALFACE
   noFill();
   stroke(255,0,0);
   for( int i=0; i<faces1.length; i++ ) {  // draw rect around detected face area(s)
       rect( faces1[i].x, faces1[i].y, faces1[i].width, faces1[i].height );
   }
 }  
  if (myCapture2.available()) {  
   myCapture2.read();  
   ocv2.copy(myCapture2);
   image(ocv2.image(),320,0);  // draw second image on the right half of the window  
   Rectangle[] faces2 = ocv2.detect();// detect anything ressembling a FRONTALFACE
   noFill();
   stroke(255,0,0);
   for( int i=0; i<faces2.length; i++ ) {
       rect( faces2[i].x + midX, faces2[i].y, faces2[i].width, faces2[i].height );
   }
 }  

}

 
public void stop() {
   ocv1.stop();
   ocv2.stop();
   super.stop();
}
 

Hello,

So — I'm also working through this problem of multiple connected web cams and Processing. If you wouldn't mind sharing your code, I'd be grateful!

You can PM me or email me julian at nearfuturelaboratory dot com

.julian.

I know this is a really old thread but i am dealing with the same exact issues and would love to take a look at your code sam_mcelhinney if thats possible.