@colortype attribute

🔌dp.kinect dp.kinect2 dp.kinect3 dp.oak

signature

colortype TYPE_STRING

values

char: 8-bit unsigned integer default
float32: 32-bit normalized floating point

examples

@colortype float32

Color component in 8-bit unsigned integer or 32-bit normalized float. This type combined with @colormap together define the memory layout of color data.

char values [0..255] are compact and often the native sensor color type
float32 values [0.0..1.0] are for computation efficiency; not increased accuracy

The plugin can convert sensor data to supported types efficiently. Consider how the data is later consumed to choose the best data type. For example, the memory size of a float32 jitter matrix is four times larger than char. Yet, having the data as float32 may speed later processing with OpenGL, filters, shaders, etc.