in the dx11 pack, there are two nodes, called RGBDepth (Kinect2) and DepthRGB (Kinect2).
Both say they do the same thing:
Returns a G32R32F formatted texture whose pixels represent a UV map mapping pixels from depth to color space.
Except that RGBDepth also says
Enable Relative Lookup to use it as displacement texture.
Now - do they map in opposite directions or do they do the same? And what’s ‘Relative Lookup to use it as displacement texture’?
Also, can I use those nodes to map from the IR image to the RGB image? Hence, does the Depth texture reliably align to the IR texture?
i think you need the world node to do the align-lookup
I want to map from one texture, the rgb image, to another texture, the IR image. World will give 3D coordinates that correspond to the IR (or depth) image space.
Found out. I’ll build a helppatch for this when I’ve got the time.
RGBDepth contains information about every pixel in the Depth or IR texture, telling us where the corresponding pixel in the RGB texture is.
DepthRGB contains infromation about every pixel in the RBG texture, telling us where the corresponding pixel in the IR and Depth textures are.
World does a very different thing, it gives you a 3D position for each pixel in the depth texture.
maybe FilterAllToRGB.tfx can support your research
greetings from Berlin to London
wish you a good Businessstart
This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.