- 135 Downloads
A physically plausible reconstruction unique up to a single unknown scale factor can be obtained from just two different images of the same scene. The differences between the two images contain enough information about the camera position and about the location of the scene points relative to the camera to make the reconstruction possible. The scale factor cannot be recovered from the images alone. It is impossible to tell if the camera is near to a small object or far away from a large object. It is also possible to reconstruct a scene up to a single unknown scale factor using the velocities in an image taken by a moving camera.
KeywordsComputer Vision Grey Level Projective Space Camera Calibration Algebraic Surface
Unable to display preview. Download preview PDF.
- Faugeras O.D. 1990 On the motion of 3D curves and its relation to optical flow. Proc. First European Conference on Computer Vision (ed. O. Faugeras). Lecture Notes in Computer Science, vol. 427, 107–117. Berlin, Heidelberg and New York: Springer-Verlag.Google Scholar
- Gibson J.J. 1950 The Perception of the Visual World. Boston: Houghton Mifflin.Google Scholar
- Harris C. 1990 Resolution of the bas-relief ambiguity in structure-from-motion under orthographic projection. BMVC90 Proc. British Machine Vision Conference, Oxford, 67–77.Google Scholar
- Horn B.K.P. 1986 Robot Vision. Cambridge, Massachusetts: The MIT Press.Google Scholar
- Marr D. 1982 Vision. San Francisco: W. H. Freeman & Co.Google Scholar
- Moffitt F.H. & Mikhail E.M. 1980 Photogrammetry (3rd edition). New York: Harper and Row.Google Scholar
- Murray D.W. & Buxton B.F. 1990 Experiments in the Machine Interpretation of Visual Motion. Cambridge, Massachusetts: The MIT Press.Google Scholar
- Semple J.G. & Kneebone G.T. 1953 Algebraic Projective Geometry. Oxford: Clarendon Press (reprinted 1979).Google Scholar
- Wolf P.R. 1983 Elements of Photogrammetry (2nd edition). McGraw-Hill.Google Scholar