Abstract: An approach is proposed for multiple-view 3D capturing by a single-view 3D camera using a simple revolve device. In this article we discuss the structure of a multiple-view 3D camera based on our method, the configure calibration for the revolve device and the alignment of the object surface 3D data captured from different views directly through space coordinate transformation. We have also demonstrated the space coordinate transformation function with regards to the revolve angle. The alignment error caused by the revolve device is analyzed, and an example is given and discussed. The multiple-view 3D camera based on our method will have many advantages:simple system, no additional operation and special requirement when capturing, computable alignment precision, automatic and rapid capturing and alignment processing.

Key words: 3D image, alignment, 3D camera, revolve device, image processing