0

[背景] 我正在帮助一名研究生(MFA)完成一个艺术项目:

我正在尝试控制 depthMap 生成的输出颜色,以便可以独立控制 Red、Green 和 Blue 值。理想的最终结果是,每个 depthMap 生成的值都可以更改为各种深浅的洋红色 (255,0,255),这样每个 SimpleOpenNI 用户就会出现差异化。

到目前为止,我还没有找到在 pde 文件中执行此操作的方法,也没有找到修改 SimpleOpenNI 类以适应相同目标的正确方法。- 无论出于何种原因,tint() 调用都会更改画布的 alpha 值,但似乎不会调整颜色值。

任何适用的建议将不胜感激。

资源:SimpleOpenNI 源 我的 pde:

import SimpleOpenNI.*;


SimpleOpenNI  context;

void setup()
{
  context = new SimpleOpenNI(this);

  // enable depthMap generation
  if(context.enableScene() == false)
  {
     println("Can't open the sceneMap, maybe the camera is not connected!"); 
     exit();
     return;
  }

  frameRate(30);
  background(128);
  size(context.sceneWidth() , context.sceneHeight()); 
}

void draw()
{

  // update the cam
  context.update(); 

  // draw irImageMap && Sets frame size.
  image(context.sceneImage(),25,25, 590,430);

  // "blend" shapes BG & previous the frame
   tint(#ED145B, 125);

}
4

1 回答 1

0

除了sceneImage(),SimpleOpenNI 还提供了一个名为的方法,该方法sceneMap()以 int[] 作为输入。它使用与该用户对应的数字填充数组中映射到特定用户的值。

例如,如果您没有用户,sceneMap 数组将用零填充(例如[0,0,0.....0]),如果您有 1 个用户,则某些元素将用 1 填充,因为属于该用户的像素(例如[0,0,0,1,1,...1,0,0,0,1,0...etc.])等等。

您可以尝试在 PImage 中进行一些颜色替换,但这不是很灵活(如果您也启用了深度图怎么办?)并且您依赖于每个用户的硬编码值(R、B、G、C ,M,Y,无论顺序可能在 SimpleOpenNI 中),因此sceneMap()是更灵活的选项。

这是一个关于如何使用它的注释示例:

import SimpleOpenNI.*;

SimpleOpenNI  context;

int[] sceneMap;//this will store data about the scene (bg pixels will be 0, and if there are any users, they will have the value of the user id - e.g. if there are no users, the array will be filled with zeros, if there is one user, some array entries will be equal to 1, etc. the size of the array is the same as the number of pixels in scene image, so it's easy to use with the pixels[] of a PImage
PImage myUserImage;//this is where we'll draw the user
int user1Colour = color(180,130,30);//change to whatever you like

void setup()
{
  context = new SimpleOpenNI(this);
  context.enableScene();

  background(200,0,0);
  size(context.sceneWidth() , context.sceneHeight());
  //set scene map array
  sceneMap = new int[context.sceneWidth()*context.sceneHeight()];
  //create the image to draw the user into, by default it will be filled black
  myUserImage = createImage( context.sceneWidth() , context.sceneHeight(), RGB );
}

void draw()
{
  context.update(); 

  // // gives you a label map, 0 = no person, 0+n = person n - tell OpenNI to update the numbers in the array
  context.sceneMap(sceneMap);
  //clear myUserImage - fill everything with black
  Arrays.fill(myUserImage.pixels,color(0));//We've never used Arrays.fill() before, but all it does is it loops through all elements on an array you pass and sets a value you want for each element - fills an array with a value
  for(int i = 0 ; i < myUserImage.pixels.length; i++){
    //check if there is a user for the current pixel, if so, use our custom colour for the pixel at this index
    if(sceneMap[i] > 0) myUserImage.pixels[i] = user1Colour;
  }
  myUserImage.updatePixels();

  //display image
  image(myUserImage,0,0);
}

从长远来看,如果你想满足多个用户的需求,编写一个基本的实用程序类是值得的:

import SimpleOpenNI.*;

SimpleOpenNI  context;

SceneMapper sceneMap;

void setup()
{
  context = new SimpleOpenNI(this);
  context.enableScene();

  background(200,0,0);
  size(context.sceneWidth() , context.sceneHeight());

  sceneMap = new SceneMapper(context);
}

void draw()
{
  context.update(); 
  sceneMap.update();
  //display image
  image(sceneMap.scene,0,0);
}
class SceneMapper{
  PImage scene;//this is a PImage where we'll actually draw the user with what colour we want
  int[] sceneMap;//this will store scene data - an array of ints which has the same length as context.sceneImage().pixels, the only difference is, sceneImage already has colours set, while scene map has numbers representing user(1,2,3etc.) on top of background(0s)
  int numPixels;//total number of pixels, we only store it so we can reuse it
  color bg;//background colour
  color[] users = {color(255),color(192),color(127),color(64),color(32)};//fill colours for users

  SceneMapper(SimpleOpenNI context){
    numPixels = context.sceneWidth()*context.sceneHeight();
    sceneMap = new int[numPixels];//init scene nap array
    scene = createImage( context.sceneWidth(), context.sceneHeight(), RGB );//create a PImage to display scene data
    scene.loadPixels();  
    bg = color(0);
  }
  void update(){
    context.sceneMap(sceneMap);//ask SimpleOpenNI to store scene map data into our array
    Arrays.fill(scene.pixels, bg);//clear the image - fill it with the background colours
    for(int i = 0 ; i < numPixels ; i++){//loop through all pixels
      for(int u = 0 ; u < users.length; u++){//loop through user colours
        if(sceneMap[i] > 0) scene.pixels[i] = users[u];//if there are user pixels, use set user colour for those pixels (e.g. pixels with value 1 will use colours stored in users[0], pixels with value 2 will use colour from users[1], etc.) 
      } 
    }
    scene.updatePixels();//we've use pixels, so update the image at the end
  }
}

但这取决于学生是否熟悉课程,并且考虑到它是 MFA,我认为清洁代码的压力较小,而完成任务的压力更大。

于 2012-05-03T19:13:45.010 回答