3D Photography from Photographs and Video Clips

This paper addresses the problem of acquiring realistic visual models of the shape and appearance of complex three-dimensional (3D) scenes from collec- tionsofimages,aprocessdubbed 3D photography. Wefocusonthreeinstances of this problem: (1) the image-based construction of projective visual hulls of complex surfaces from weakly-calibrated photographs; (2) the automated matching and registration of photographs of textured surfaces using a-ne- invariant patches and their geometric relationships; and (3) an approach to projectivemotionanalysisandself-calibrationexplicitlyaccountingfornatural cameraconstraintssuchaszeroskewandcapableofhandlinglargenumbersof imagesinane-cientanduniformmanner. Wealsobrie∞ydiscusssomerelat- edapplicationsoforienteddifierentialprojectivegeometrytocomputervision problems,includingthedeterminationoftheorderingofrimsegmentsinpro- jectivevisualhullcomputation,andapurelyprojectiveproofofKoenderink's famous characterization of the localshape of visual contours.