Kinect - Background removal with video

Azeb

Hello everyone ! I'm a french student in communication and i have a problem : I need to use the kinect to remove a background behind someone, and put a video (in loop) as new background :) I'm working with processing 2.1.1. and SimpleOpenNI Version 1.96

I find this code, who remove the background and put an image, but i need a video :/

If someone has any idea ! :)

import processing.opengl.*;
import SimpleOpenNI.*;
//install this on your machine and also the library in processing
//http://code.google.com/p/simple-openni/wiki/Installation


SimpleOpenNI kinect;
//based on Greg's Book Making things see.
boolean tracking = false;
int userID; int[] userMap;
// declare our images
PImage backgroundImage;
PImage resultImage;
void setup() {
  size(640*2, 480);
  // load the background image
  backgroundImage = loadImage("http://www.eheart.com/SHASTA/wallpaper/alpenglow-640x480.jpg");
  kinect = new SimpleOpenNI(this);
  if(kinect.isInit() == false)
  {
     println("Can't init SimpleOpenNI, maybe the camera is not connected!");
     exit();
     return;
  }

  // enable depthMap generation
 kinect.enableDepth();

  // enable skeleton generation for all joints
  kinect.enableUser();
  // enable color image from the Kinect
  kinect.enableRGB();
  //enable the finding of users but dont' worry about skeletons

  // turn on depth/color alignment
  kinect.alternativeViewPointDepthToImage();
  //create a buffer image to work with instead of using sketch pixels
  resultImage = new PImage(640, 480, RGB);
}
void draw() {
  kinect.update();
  // get the Kinect color image
  PImage rgbImage = kinect.rgbImage();

  image(rgbImage, 640, 0);
  if (tracking) {
    //ask kinect for bitmap of user pixels
    loadPixels();
    userMap = kinect.userMap();
    for (int i =0; i < userMap.length; i++) {
      // if the pixel is part of the user
      if (userMap[i] != 0) {
        // set the pixel to the color pixel
        resultImage.pixels[i] = rgbImage.pixels[i];
      }
      else {
        //set it to the background
        resultImage.pixels[i] = backgroundImage.pixels[i];
      }
    }

    //update the pixel from the inner array to image
     resultImage.updatePixels();
    image(resultImage, 0, 0);
  }
}


void onNewUser(SimpleOpenNI curContext, int userId)
{
 userID = userId;
  tracking = true;
  println("tracking");
  //curContext.startTrackingSkeleton(userId);
}

void onLostUser(SimpleOpenNI curContext, int userId)
{
  println("onLostUser - userId: " + userId);
}

void onVisibleUser(SimpleOpenNI curContext, int userId)
{
  //println("onVisibleUser - userId: " + userId);
}

Love,

Azeb

blyk

You can try this but I am not sure whether it would work or not. I don't have Kinect so I can test it by myself so tell me if you get any success.

import processing.opengl.*;
import processing.video.*; // I have added : importing processing video library
import SimpleOpenNI.*;
//install this on your machine and also the library in processing
//<a href="http://code.google.com/p/simple-openni/wiki/Installation" target="_blank" rel="nofollow">http://code.google.com/p/simple-openni/wiki/Installation</a>;


SimpleOpenNI kinect;
Movie mov; // I have added: Creating movie object
//based on Greg's Book Making things see.
boolean tracking = false;
int userID; 
int[] userMap;
// declare our images
PImage backgroundImage;
PImage resultImage;

void movieEvent(Movie m) { // movie reading
  m.read();
}

void setup() {
  size(640*2, 480, P3D); // use P3D with size. It will make your program a bit faster becuase it uses system GPU for calculation and graphics rendering
  // load the background image
  //  backgroundImage = loadImage("Test.jpg"); // I have commented it becuase you don't want it
  // 
  mov = new Movie(this, "transit.mov"); // I have added: load a movie file  
  kinect = new SimpleOpenNI(this);
  if (kinect.isInit() == false)
  {
    println("Can't init SimpleOpenNI, maybe the camera is not connected!");
    exit();
    return;
  }

  // enable depthMap generation
  kinect.enableDepth();

  // enable skeleton generation for all joints
  kinect.enableUser();
  // enable color image from the Kinect
  kinect.enableRGB();
  //enable the finding of users but dont' worry about skeletons

  // turn on depth/color alignment
  kinect.alternativeViewPointDepthToImage();
  //create a buffer image to work with instead of using sketch pixels
  resultImage = new PImage(640, 480, RGB);
}
void draw() {
  kinect.update();
  // get the Kinect color image
  mov.play(); // I have added: play movie 
  PImage rgbImage = kinect.rgbImage();
  //  image(rgbImage, 0, 0);
  image(mov, 0, 0); 
  if (tracking) {
    //ask kinect for bitmap of user pixels
    loadPixels();
    resultImage.userMap = kinect.userMap();
    for (int i =0; i < userMap.length; i++) {
      // if the pixel is part of the user
      if (userMap[i] != 0) {
        // set the pixel to the color pixel
        resultImage.pixels[i] = rgbImage.pixels[i];
      }
      else {
        //set it to the background
        // resultImage.pixels[i] = backgroundImage.pixels[i];
        resultImage.pixels[i] = mov.pixels[i];
      }
    }

    //update the pixel from the inner array to image
    resultImage.updatePixels();
    image(resultImage, 0, 0);
  }
}


void onNewUser(SimpleOpenNI curContext, int userId)
{
  userID = userId;
  tracking = true;
  println("tracking");
  //curContext.startTrackingSkeleton(userId);
}

void onLostUser(SimpleOpenNI curContext, int userId)
{
  println("onLostUser - userId: " + userId);
}

void onVisibleUser(SimpleOpenNI curContext, int userId)
{
  //println("onVisibleUser - userId: " + userId);
}