Hi there folks, have recently purchased an Azure and been going through tutorials for pointcloud manipulation, hoping to do some installation work in the future.
I’ve noticed gigantic black shadows while rendering, like I’m getting my pointcloud input plus a black 2D matte after pulling into a Geo COMP. Have included images and my network, nothing too complicated. Wondering if there’s a setting that I’ve missed, or if there’s a straightforward way to make these pointclouds a tad more usable?
Not sure what you mean. Yould you maybe mark the shadows that you mean.
The kinect is still using light to a pointSource to calculate depth, so things will throw an shadow as you cannot move like arround a solid object (we are not there yet with artifical gravity fields.)
x2, I don’t understand exactly what you mean either. In the first screenshot I can see data that is out of range for the sensor from what seems to be a window. This is expected behavior: Out of range, shiny, transparent or otherwise refractive / reflective objects will cause the depth readings to fail and will appear as zero-valued data.
In the second screenshot I see a person, a desk and a wall, in that wall there appears to be a shadow of this person’s head. This is also expected behavior, the sensor ‘sees’ depth just like an optical camera, meaning that it can only see what is in front of it.
