Abstract

In the general case, a trilinear relationship between three perspective views is shown to exist. The trilinearity result is shown to be of much practical use in visual recognition by alignment --- yielding a direct reprojection method that cuts through the computations of camera transformation, scene structure and epipolar geometry. Moreover, the direct method is linear and sets a new lower theoretical bound on the minimal number of points that are required for a linear solution for the task of reprojection. The proof of the central result may be of further interest as it demonstrates certain regularities across homographies of the plane and introduces new view invariants. Experiments on simulated and real image data were conducted, including a comparative analysis with epipolar intersection and the linear combination methods, with results indicating a greater degree of robustness in practice and a higher level of performance in re-projection tasks. Keywords--- Visual Recognition, Al...

Citations

545	A computer algorithm for reconstructing a scene from two projections - Longuet-Higgins - 1981
460	What can be seen in three dimensions with an uncalibrated stereo rig - Faugeras - 1992
372	The interpretation of visual motion - Ullman - 1979
366	Recognition by linear combinations of models - Ullman, Basri - 1991
270	Uniqueness and estimation of three dimensional motion parameters of rigid objects with curved surfaces - Tsai, Huang - 1984
267	Affine structure from motion - Koenderink, Doorn - 1990
221	Stereo from uncalibrated cameras - Hartley, Gupta, et al. - 1992
221	Recognizing solid objects by alignment with an image - Huttenlocher, Ullman - 1990
171	Canonic representations for the geometries of multiple projective views - Luong, Vieville - 1993
136	Aligning pictorial descriptions: an approach to object recognition - Ullman - 1989
135	Layered representation for motion analysis - Wang, Adelson - 1993
114	Inherent ambiguities in recovering 3d motion and structure from a noisy flow field - Adiv - 1989
113	Relative orientation - Horn - 1990
100	Trilinearity in Visual Recognition by Alignment - Shashua - 1994
99	What can two images tell us about the third one - Faugeras, Robert - 1996
82	On determining the Fundamental matrix: analysis of different methods and experimental results - Luong, Deriche, et al. - 1993
67	Geometry and Photometry in 3D Visual Recognition - Shashua - 1992
56	Projective structure from uncalibrated images: Structure from motion and recognition - Shashua - 1994
54	Hierarchical motion-based frame rate conversion - Bergen, Hingorani - 854
52	Relative affine structure: Theory and applications to 3d reconstruction from perspective views - Shashua, Navab - 1994
40	Factoring image sequences into shape and motion - Tomasi, Kanade - 1991
39	Projective Depth: A Geometric Invariant for 3D Reconstruction From Two PerspectivenOrthographic Views and For Visual Recognition - Shashua - 1993
32	Relative Orientation Revisited - Horn - 1991
28	Affine and Projective Structure from Motion - Demey, Zisserman, et al. - 1992
27	The reconstruction of a scene from two projections: configurations that defeat the 8-point algorithm - Longuet-Higgins - 1984
24	Invariant linear methods in photogrammetry and modelmatching - Barrett, Brill, et al. - 1992
24	The projective geometry of ambiguous surfaces - Maybank - 1990
24	Projective geometry for machine vision - MUNDY, A - 1992
22	On Geometric and Algebraic Aspects of 3D Affine and Projective Structures from Perspective 2D Views - Shashua - 1993
19	Layered representation for image coding - Adelson - 1991
19	Correspondence and Affine Shape from two Orthographic Views: Motion and Recognition - Shashua - 1991
19	Illumination and view position in 3d visual recognition - Shashua
16	A unified perspective on computational techniques for the measurement of visual motion - Anandan - 1987
13	Contour Matching Using Local Affine Transformations - Bachelder, Ullman - 1992
12	The quadric reference surface: Applications in registering views of complex 3d objects - Shashua, Toelg - 1994
7	3-D model alignment without computing pose - Mundy, Welty, et al. - 1992
6	Model based invariants for 3-D vision - Weinshall - 1993
5	Robustness of correspondence based structure from motion - Dutta, Synder - 1990
4	stereo vision is not always about 3D reconstruction - Why - 1993