3dpalign is a tool for use in a 3D photography system. It provides a robust, easy-to-use implementation of user-assisted registration of multiple range images.

3dpalign was conceived, designed and written by Peter Schroeder (ps@cs.caltech.edu), Daniel Azuma (dazuma@kagi.com), and Bradley Nelson (bdnelson@ugcs.caltech.edu) with the Multi-Res Modeling Group at Caltech. It was designed around, and optimized for use in the greater 3D photography system being developed by the Multi-Res Modeling Group.

3dpalign is designed on the theory that humans have a knack for feature identification, but are poorly skilled at actual manual manipulation of three-dimensional objects. Therefore, the user is merely required to identify corresponding features in the overlapping areas of adjacent range images, through a point-and-click interface. Once this has been done for enough pairs of adjacent range images, the system can find a good global solution for the alignment. 3dpalign sports an easy-to-use graphical user interface driven by mouse actions and menus.