View my account

how to render depth data like realsenser view

Comments

11 comments

  • MartyG

    The RealSense Viewer has a range of post-processing filters enabled by default that improve the image quality.  These can be found in the Post-Processing section of the options at the side of the RealSense Viewer window.

    When you create your own application though, you have to program in the post-processing filters.  So there can be a difference between the output of the Viewer and your own application.

    The link below has a C++ tutorial for adding post-processing filters.

    https://github.com/IntelRealSense/librealsense/tree/master/examples/post-processing 

    And here is a tutorial for Python.

    https://github.com/IntelRealSense/librealsense/blob/jupyter/notebooks/depth_filters.ipynb 

    0
    Comment actions Permalink
  • Fred Cao

    Hello MartyG:

            Thanks for your remind.

            It is useful for me.

            After Study RealSense Viewer. I get the key point. 

            It is "Visual Preset:" setting. from the RealSense  Viewer It is clear, I put compare picture under.

            I would like to use  this feature in  my app, Would you kindly tell me which API or function responsible this feature.

    Best Regards

    Fred

           

    0
    Comment actions Permalink
  • MartyG

    What language are you using please Fred?  C++, C# or Python?  Thanks!

    0
    Comment actions Permalink
  • Fred Cao

    C++

    0
    Comment actions Permalink
  • MartyG

    Unfortunately the C++ code for setting this particular option does not seem to be simple.  It is covered by lines 172 to 210 of the SDK file colorizer.cpp

    https://github.com/IntelRealSense/librealsense/blob/master/src/proc/colorizer.cpp#L172 

     

     

    0
    Comment actions Permalink
  • Fred Cao

    Hi MartyG:

           I read the sourc code, I fould it is new version at github  compare with install location.

           If I try to integrate this feature into old version It will take me more time.

           I try to directly use the new version.

          Unfortunately, after download the new version. there is no *.sln VS project files for samples, only makefiles.

          and read the readme.md file, I try to use vcpkg.exe, it looks not working.

          Could you kindly tell me how the build new realsenser version from github(include lib,examples).

          

     

    Best Regards

    Fred 

    0
    Comment actions Permalink
  • MartyG

    If you would prefer to build Librealsense using vcpkg, instructions can be found here:

    https://github.com/IntelRealSense/librealsense/blob/master/readme.md#building-librealsense---using-vcpkg 

     

    0
    Comment actions Permalink
  • Fred Cao

    Hi MartyG:

              It is failed for me to use vcpkg. fortunately I could use cmake to create VS sln project.

             When I try to use colorizer.cpp, I refer RealSenser Viewer, It is hard. :(

             Is there any simple example?

     

    Best Regards

    Fred Cao 

    0
    Comment actions Permalink
  • MartyG

    The tutorial for the rs-capture example program explains how the colorizer can be applied to a image.

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

    0
    Comment actions Permalink
  • Fred Cao

    Hi MartyG:
    Thanks for your hint. It is helpful for me.
    Now Find find 2 colorizer, one is in rs2 namespace. another is in librealsense
    To support Dynamic,Fixed,Near,Far feature, it need implment librealsense colorizer.
    But strange I could not find this colorizer used in RealSenser Viewer.
    Would you kindly tell me if my understand correct? and where could I find librealsense colorizer used in viewer?
    Best Regards
    Fred

    0
    Comment actions Permalink
  • MartyG

    I hope that the colorizer code in the link below will solve your problem with these settings:

    https://github.com/IntelRealSense/librealsense/issues/5371 

     

    0
    Comment actions Permalink

Please sign in to leave a comment.