2021-06-23_11-49-45_Image To Material

The Image to Material template allows to generate a high-quality PBR material from a single input image.

This template has two main algorithms:

  • AI Powered
  • B2M

See below for a detailed explanation of each algorithm.


Here is an example of material channels generated from a single input image:


To change the algorithm of the Image to Material template, click on the dropdown below the template name:

AI Powered

The AI Powered algorithm uses machine learning to recognize shapes and objects and generate accurate height and normal information. This algorithm also get rid of any shadows and highlight from the albedo. 

The neural network has been only trained on outdoor materials at the moment. Other kind of materials may produce unexpected results.

This template requires specific hardware. It is available only on Windows/Linux and with an Nvidia GPU. See technical requirements.


The B2M algorithm uses the Substance based Bitmap to Material method to generate multiple channels such as base color, normal, metallic, roughness, and ambient occlusion using procedural techniques.

This algorithm may produce less accurate results but will work on a wider range of input images.