Physical model is a common tool for testing of the product features during the design process. This model is usually made of clay or plastic because of the modifiability of these materials. Therefore, the designer could easily adjust the model shape to enhance the look or ergonomics of the product. Nowadays, some companies use augmented reality to enhance their design process. This concept is called augmented prototyping. Common approach uses artificial markers to augment the product prototype by digital 3D models. These 3D models that are shown on the markers positions can represent e.g. car spare parts such as different lights, wheels, spoiler etc. This allows the designer interactively change the look of the physical model. Further, it is also necessary to transfer physical adjustments made on the model surface back to the computer digital model. Well-known tool for this purpose a professional 3D scanner. Nevertheless, the cost of such scanner is substantial. Therefore, we focused on different solution – a motion capture device Microsoft Kinect that is used for computer games. This article outlines a new augmented prototyping approach that directly updates the digital model during the design process using Kinect depth camera. This solution is a cost effective alternative to the professional 3D scanners. Our article describes especially how depth data can be obtained by the Kinect and also provides an evaluation of depth measurement precision.