A 3D Authoring Method by Editing Real World Scene

In recent years, consumer 3D printers and depth cameras become more popular input/output devices in the field of 3D modeling and its applications. A high barrier for the practical use of the technology is 3D geometric data production. Although many 3D authoring tools are available for personal computer users, appropriately designing 3D geometry still is a difficult task. We propose a method for easily creating new 3D geometric models by using live depth images. In our approach, a 3D geometric data captured by a consumer depth camera such as Kinect is used for creating new objects. Our system allows a user to extract a specific part in the whole captured data with a simple cut off operation and combine such extracted 3D objects for making a new model. The created 3D geometry can be formed by using a 3D printer. A lot of professional 3D modeling systems such as Blender, 3DS MAX and MAYA provide similar functions, but all of these systems are very hard to master basic operations especially by novices. The proposed system enables even novices to perform their own 3D model design and validate it by only using consumer devices.