View my account

Processing inside the Camera internal processor

Comments

3 comments

  • MartyG

    Hi Dhruvdarda2001  It is not possible to perform the post-processing on the camera hardware, unfortunately.  The camera does not have a CPU and only has 16 mb storage space in its EEPROM flash memory component for storing the firmware driver.

     

    RealSense cameras do not obtain the depth frame from the RGB image.  It is instead calculated using raw left and right infrared images (not the Infrared and Infrared 2 streams) in the camera hardware.  Because left and right infrared frames are used, this is why the camera's technology is called stereo depth.

     

    As you are using C++ code, you could potentially improve performance by using the SDK's GLSL Processing Blocks system to offload work from the CPU to a graphics GPU on the computer.  GLSL is 'vendor neutral', meaning that it should work with any GPU brand including built-in ones, though the difference in performance may not be noticable on low-end computers / computing devices.

     

    The link below has a very good pros and cons analysis of GLSL processing blocks and when it can be used.

    https://github.com/IntelRealSense/librealsense/pull/3654

     

    There is also a C++ rs-gl example program.

    https://github.com/IntelRealSense/librealsense/tree/master/examples/gl

    0
    Comment actions Permalink
  • Dhruvdarda2001

    Thank you @MartyG. This helps!
    I am actually using the Python API, so the same thing would be available there as well, right?

    0
    Comment actions Permalink
  • MartyG

    GLSL would be problematic for Python because its implementation involves editing C++ RS2:: script instructions to RS2::GL:: but pyrealsense2 does not use RS2:: commands.

     

    The other way to offload processing from CPU to GPU is to use an Nvidia Jetson board as the computer, as the RealSense SDK's CUDA support can then be enabled to automatically GPU-accelerate alignment and pointclouds without having to edit any code.

    1
    Comment actions Permalink

Please sign in to leave a comment.