OpenCV  4.6.0-dev
Open Source Computer Vision
Using Creative Senz3D and other Intel RealSense SDK compatible depth sensors

Prev Tutorial: Using Orbbec Astra 3D cameras

realsense.jpg
hardwares

Note: This tutorial is partially obsolete since PerC SDK has been replaced with RealSense SDK

Depth sensors compatible with IntelĀ® RealSense SDK are supported through VideoCapture class. Depth map, RGB image and some other formats of output can be retrieved by using familiar interface of VideoCapture.

In order to use depth sensor with OpenCV you should do the following preliminary steps:

  1. Install Intel RealSense SDK 2.0 (from here https://github.com/IntelRealSense/librealsense).
  2. Configure OpenCV with Intel RealSense SDK support by setting WITH_LIBREALSENSE flag in CMake. If Intel RealSense SDK is found in install folders OpenCV will be built with Intel Realsense SDK library (see a status LIBREALSENSE in CMake log).
  3. Build OpenCV.

VideoCapture can retrieve the following data:

  1. data given from depth generator:
    • CAP_INTELPERC_DEPTH_MAP - each pixel is a 16-bit integer. The value indicates the distance from an object to the camera's XY plane or the Cartesian depth. (CV_16UC1)
    • CAP_INTELPERC_UVDEPTH_MAP - each pixel contains two 32-bit floating point values in the range of 0-1, representing the mapping of depth coordinates to the color coordinates. (CV_32FC2)
    • CAP_INTELPERC_IR_MAP - each pixel is a 16-bit integer. The value indicates the intensity of the reflected laser beam. (CV_16UC1)
  2. data given from RGB image generator:
    • CAP_INTELPERC_IMAGE - color image. (CV_8UC3)

In order to get depth map from depth sensor use VideoCapture::operator >>, e. g. :

VideoCapture capture( CAP_REALSENSE );
for(;;)
{
Mat depthMap;
capture >> depthMap;
if( waitKey( 30 ) >= 0 )
break;
}

For getting several data maps use VideoCapture::grab and VideoCapture::retrieve, e.g. :

VideoCapture capture(CAP_REALSENSE);
for(;;)
{
Mat depthMap;
Mat image;
Mat irImage;
capture.grab();
capture.retrieve( depthMap, CAP_INTELPERC_DEPTH_MAP );
capture.retrieve( image, CAP_INTELPERC_IMAGE );
capture.retrieve( irImage, CAP_INTELPERC_IR_MAP);
if( waitKey( 30 ) >= 0 )
break;
}

For setting and getting some property of sensor` data generators use VideoCapture::set and VideoCapture::get methods respectively, e.g. :

VideoCapture capture(CAP_REALSENSE);
cout << "FPS " << capture.get( CAP_INTELPERC_DEPTH_GENERATOR+CAP_PROP_FPS ) << endl;

Since two types of sensor's data generators are supported (image generator and depth generator), there are two flags that should be used to set/get property of the needed generator:

For more information please refer to the example of usage videocapture_realsense.cpp in opencv/samples/cpp folder.