Skip to content

3D Vision


The 3D vision is the extension of the classical computer vision. An image is made of XY coordinates and RGBA colors. In 3D vision we simply add the Z axis to work in XYZ. Adding a dimension is not neutral. Instead of considering all the points of an image, we consider all the points in a cube that are occupied by an object. The other points are usually empty. Working in 3D means working on a much larger amount of data within which there are many gaps. This is a different challenge for which has developed several strategies, which allow it today, for example, to model entire urban areas in order to identify and locate the various facilities.

3D vision has many practical applications. Some examples to get an idea:

  • Make an inventory of urban furniture (RIM3D)
  • Identify the precise position of industrial parts arranged in bulk, to enable a robot to pick them up
  • Identify the presence of people or body parts in areas that need to be secured
  • Analyze the road in front of a car
  • Precise control of part positioning using high precision 3D cameras
  • Enable human-machine collaboration by knowing who is where at all times

The 3D vision is a specific competence of, on which we have developed a real know-how.

The wide range of sensors available, from those that measure a part to within a hundredth of a millimeter to those that can cover an entire field, allows for a wide range of applications where 2D vision finds its limits.


Leave a Reply