Mitsubishi Electric Research Labs.

Remo Ziegler

Wojciech Matusik

Hanspeter Pfister

Leonard McMillan, Jr

Wai Kit Addy Ngan

A method constructs three-dimensional (3D) models of a scene from a set of two-dimensional (2D) input images. The 3D model can then be used to reconstruct the scene from arbitrary viewpoints. A user segments and labels a set of corresponding polygonal regions in each image using conventional photo-editing tools. The invention constructs the model so that the model has a maximum volume that is consistent with the set of labeled regions in the input images. The method according to the invention directly constructs the polygonal model.

Three-dimensional scene reconstruction from labeled two-dimensional images

A method models a three-dimensional object by first acquiring alpha mattes of the object for multiple viewpoints. The alpha mattes are then projected onto a surface hull completely enclosing the object to construct an opacity hull storing opacity values of the surface of the object. The object is illuminated for various lighting conditions while images are acquired. The images are projected onto the opacity hull to render the object under arbitrary lighting conditions for arbitrary viewpoints.

Growing community of inventors

Cambridge, MA, United States of America

Remo Ziegler