Over the years, we have developed several techniques for recognising and tracking 2D and 3D objects in real-time. These techniques are able to cope with variability of the lighting conditions, presence of clutter, background texture and general image noise. This is quite critical to the success of machine vision solutions in industrial environments where all those factors are difficult to control.

Using our in-house expertise, we can develop solutions for the recognition and the tracking of multiple objects in acquired image sequences, for instance in an assembly or packaging line. We can also deal with deformable shapes as well as objects with variable shape (as in the case of the food manufacturing industry). Using software libraries developed in-house, we are able to develop prototypes and final solutions in a relatively short time.