internal parameter of each camera

intr.width, intr.height, intr.fx, intr.fy, intr.ppx, intr.ppy

• MartyX Grover

  • December 26, 2021 10:07

  The intrinsic parameters of the camera are described in the Intrinsics section of the SDK's Projection documentation in the link below.

  https://dev.intelrealsense.com/docs/projection-in-intel-realsense-sdk-20#section-intrinsic-camera-parameters

   

  You can generate a point cloud without aligning depth to color with align_to by using an SDK instruction called pc.map_to(color) and then additionally obtaining the point cloud coordinates if you need to with the instruction points.get_vertices()

  Here is an example Python script for doing so:

  https://github.com/IntelRealSense/librealsense/issues/1231

   

  If you only need a 3D depth value for a single coordinate then you can use rs2_project_color_pixel_to_depth_pixel instead of aligning the entire image and generating a point cloud.

  https://github.com/IntelRealSense/librealsense/issues/6239#issuecomment-614261704

   

  I note though that you mention that you do not have a depth frame and color frame.  In that case, you may be able to simulate them using the RealSense SDK's software-device 'synthetic camera' interface, though it will work best with C++ rather than Python.  The link below has a C++ tutorial for generating a point cloud with software-device.

  https://github.com/IntelRealSense/librealsense/tree/master/examples/software-device

  0

  Comment actions Permalink

Please sign in to leave a comment.