Latest technologies clearly show an improvement in 3D data acquisition and 3D object reconstruction techniques. Therefore Kinect device has become popular not only as an appliance from Xbox console, but also for researching projects, especially in robotics and computer vision, due to its availability and high performance capturing color and depth images as well as audio.
The aim of this talk is to present the stages of these two first processes. In reconstruction, the main problem is to find out accurately correspondent points between the acquired frames. Here an improvement of correspondence problem task using a refined version of the Iterative Closest Point algorithm is applied, this upturn is able to have a faster performance compared with the original version. Finally further possible prosecutions will be discussed.