Using Image Data in Implicit Modeling

Import data from image files for use in implicit modeling workflows.

Images can contain useful data for implicit workflows such as logos, text, bump maps for textures, or heat maps of analysis results. Image data can be used to both create or modify implicit geometry. This information can be used to locally offset surfaces, change thicknesses. radii, and much more. An implicit image object is a purely 2D implicit function based on the pixel data contained in the image. The resolution of the implicit field created from the image will depend on the resolution of the image.

  1. In the Image dialog box, click the Import icon. , use the file browser to select the image (.png) file you want to import and click Open.

  2. Use the Channel dropdown to select which information should be extracted from each pixel. For a PNG file with color data, Red, Green, Blue, and Alpha channels are available. For a Grayscale PNG, only the Grayscale channel will be available.
  3. Use the Flip checkbox to invert the data in the image. The intensity data for each pixel is typically in the range [0,255]. Inverting the image is the same as subtracting the true pixel intensity value from 255. In a grayscale image, this converts white pixels to black, and vice versa.
  4. Use the Normalize checkbox to change the range of values produced in the field to a more intuitive range for use in downstream operations. Instead of using pixel intensities in the range [0,255], the ranges can be rescaled to the ranges defined below.
  5. The Convention buttons specify which range to use as the output range for the field created from image data.
    • 0 to 1 will normalize the [0, 255] pixel data to [0, 1]. This range will not create a surface as no values will be in the negative range, which is usually the convention for regions in the interior of an object. The field created from this data will therefore be an unsigned scalar field.
    • -1 to 1 will normalize the [0, 255] pixel data to [-1, 1]. A surface will be created where the normalized pixel values are 0 and the field created from the image data will be a signed scalar field.
  6. Click OK.