How it works

After loading the image, the deep learning algorithm transforms the image into a 32-bit depth map, in which the light and dark pixels contain information about the distance of the object's surface from the point of view.

color_map depth_map

This depth map information is used as a displacement map to offset vertices on the generated object plane.

The result can be downloaded in glTF, GLB or STL format and used in 3D computer graphics programs such as Blender, Autodesk Maya, Cinema 4D, etc.