3D SURFACE ESTIMATION AND MODEL CONSTRUCTION FROM SPECULAR MOTION IN IMAGE SEQUENCES

This work reconstructs a 3D graphics model of an object with specular surfaces by its rotation. Continuous images are taken to measure highl ights on the smooth surfaces and their motion. Coplanar extended light s determine a plane of rays to produce a highlight stripe on the objec t. 3D surfaces are then recovered from the moving stripe. We investiga te global motion characteristics of the highlights in the Epipolar-pla ne images so as to qualitatively identify surface types and control th e modeling process. Under single and multiple plane-of-rays illuminati ons, we give two quantitative approaches for surface and normal recove ry which use highlight orientation in the image and highlight motion i n the Epipolar-Plane images. The computations employ a first-order dif ferential equation and linear equations, respectively.