Some tricks and hints for using LTI-Lib at Montefiore. LTI-Lib is an object oriented library with algorithms and data structures frequently used in image processing and computer vision. It was developed at the RWTH-Aachen University as a part of many research projects on computer vision dealing with robotics, object recognition, sign language, and gesture recognition. It provides an object oriented C++ library that includes fast algorithms, which can be used in real applications.

Pr. Piater's webpage). Then, it is very easy :

".@$highlighter->highlight_text("  
#include \"VideoClient.h\"

/* Declare the videoClient */
VideoClient vc(true);
  /* Open it */  
vc.open(0,0); 

/* Define a LTI-Lib image*/
image img;     

/* Loop over the frames */
for (;;) {

/* Get the pointer to the image data in shared memory */
image_ptr = (rgbPixel *)vc.getFrameBuffer();
/* Associates img to the data from image_ptr */
img.useExternData(vc.getFrameHeight(),vc.getFrameWidth(),image_ptr);
/*
* Your processing goes here
* ...
*/
/* Asks for next frame and blocks it */
vc.nextFrame(true,true,true,false);
}
");


$mailinglist = "Whenever you are facing a problem, you have a lot of solutions to help you :


  
 RTFM:  LTI-Lib documentation 
   
  LTI-Lib Wiki 
   
 
LTI-Lib 
funs, by Toru Tamaki, with a lot of example code, tips, etc.
  
 Mail/phone/come and see me.
 
 Use the  LTI-Lib mailing list  at SourceForge. Very kind people will try to give answers to all our problems.
 This is not directly linked to LTI-Lib but may be useful if you want to build your videos from your program :   libavcodec 
 To debug \"à la purify\" in linux :  Valgrind  
 
  

"; $tmpvar3 = " Use libavcodec/libavformat from the ffmpeg API !
I made a small class that do that and which is very easy to use. Mail me if interested.
"; $tmpvar4 = " First, install the Bayesian Filtering Library .
Here is a small program that implements a very naive color histogram tracker with particle filter.

• Here is the source code
• This movie is an example of the output

"; $tmpvar5 = " Yes, by specifying the shm keys for each svideod.

 svideod-mpeg_play -k 0 -a toto.mpeg &
 svideod-mpeg_play -k 1 -a tata.mpeg &

Then the code that reads the images could be:

".@$highlighter->highlight_text("
  lti::image img1, img2;
  lti::rgbPixel *image_ptr1,*image_ptr2;
  // Declare the videoClients 
  VideoClient vc1(true),vc2(true);
  // Open them, second argument is the shm key
  vc1.open(0,0);
  vc2.open(0,1);
  viewer v1,v2;
  
  for  (;;){
   
    // Get the pointer to the image data in shared memory 
    image_ptr1 = (lti::rgbPixel *)vc1.getFrameBuffer();
    image_ptr2 = (lti::rgbPixel *)vc2.getFrameBuffer();

    // Associates img to the data from image_ptr 
    img1.useExternData(vc1.getFrameHeight(),vc1.getFrameWidth(),image_ptr1);
    img2.useExternData(vc2.getFrameHeight(),vc2.getFrameWidth(),image_ptr2);

    // Visualize
    v1.show(img1);
    v2.show(img2);

    // Asks for next frame and blocks it 
    vc1.nextFrame(true,true,true,false);
    vc2.nextFrame(true,true,true,false);
  }");

$tmpvar4 = " First, install the  Bayesian Filtering Library .

Here is a small program that implements a very naive color histogram tracker with particle filter.

  
  Here  is the source code
  
 This  movie  is an example of the output

 

"; $tmpvar6 = "For users who work with the LTI-Lib on Windows platforms might like to take a look at Impresario . This is a small Rapid Prototyping System which supports the development and evaluation of simple image processing systems and incorporates the LTI-Lib into an extendable graphical user interface.It has been developped at RTWH. "; ?>

") ; ?>

") ; ?>