what is encoded by the colors in the compute principal curvature directions filter

Question

I am new to to meshlab and I have a question about how information is encoded in the color maps generated by the compute curvature principal directions filter. The parameters for this filter include several different measures of curvature (e.g. mean, gaussian, k1, etc) which are all independent of the principal directions, and the output of the filter is a scaler color map. I would be most appreciative if someone could explain what precisely is encoded by those colors

Answer 1

Start by reviewing this answer for better understanding of curvatures.

Principal curvatures on each vertex are encoded as two colored vectors:

• red for maximal curvature, or k1
• green for minimal curvature, or k2

The median, Gaussian, k1 and k2 are stored as scalar values in the quality of each vertex. The value is converted to colours according to your configuration in the quality mapper. By default red represent the minimum global value of curvature on the mesh, and blue the maximum global. So colour distribution depends of the difference between min and max values on the mesh. You can use the menu option Render->Show Quality Histogram to easily view the asignation of values to colors.

See this answer to see an example of how to change the colour applied to values using the quality mapper..