Abstract

This thesis describes some new approaches to the representation and analysis of visual motion, as perceived by a biological or machine visual system. We begin by discussing the computation of image motion fields, the projection of motion in the three-dimensional world onto the two-dimensional image plane. This computation is notoriously difficult, and there are a wide variety of approaches that have been developed for use in image processing, machine vision, and biological modeling. We show that a large number of the basic techniques are quite similar in nature, differing primarily in conceptual motivation, and that they each fail to handle a set of situations that occur commonly in natural scenery. The central theme of the thesis is that the failure of these algorithms is due primarily to the use of vector fields as a representation for visual motion. We argue that the translational vector field representation is inherently impoverished and error-prone. Furthermore, there is evidence that a ...

Citations

1480	An iterative image registration technique with an application to stereo vision - Lucas, Kanade - 1981
1273	Robot Vision - Horn - 1986
715	Vision: A Computational Investigation into the Human Representation and Processing of Visual Information - MARR - 1982
634	Theory of communication - Gabor - 1946
459	Spatiotemporal energy models for the perception of motion - Adelson, B - 1985
453	Uncertainty relation for resolution in space, spatial frequency, and orientation optimized by twodimensional visual cortical filters - Daugman - 1985
401	A computational framework and an algorithm for the measurement of visual motion - Anandan - 1989
301	The perception of the visual world - Gibson - 1950
271	Computation of component image velocity from local phase information - Fleet, Jepson - 1990
214	Determining three-dimensional motion and structure from optical flow generated by several moving objects - Adiv - 1985
187	Normalization of cell responses in cat striate cortex - HEEGER - 1992
166	An investigation of smoothness constraints for the estimation of displacement vector fields from image sequences - Nagel, Enkelmann - 1986
143	Model of human visual-motion sensing - Watson, Ahumada - 1985
133	On the computation of motion from sequences of images - a review - Aggarwal, Nandhakumar - 1988
128	Scene segmentation from visual motion using global optimization - Murray, Buxton - 1987
112	Deformable kernels for early vision - Perona - 1995
109	Phenomenal coherence of moving visual patterns - Adelson, Movshon - 1982
107	Velocity determination in scenes containing several moving images - Fennema, Thompson - 1979
106	Functional properties of neurons in middle temporal visual area of the Macaque monkey. I. Selectivity for stimulus direction - Maunsell, Essen - 1983
103	Autocorrelation, a principle for the evaluation of sensory information by the central nervous system - Reichardt - 1961
102	Displacement Vectors Derived from Second Order Intensity Variations - Nagel - 1983
100	Image analysis using multigrid relaxation methods - Terzopoulos - 1986
86	Robust dynamic motion estimation over time - Black, Anandan - 1991
85	Modeling and estimation of multiresolution stochastic processes - Basseville, Benvensite, et al. - 1992
84	Robust estimation of a multi-layered motion representation - Darrell, Pentland - 1991
84	Directional selectivity and its use in early visual processing - Marr, Ullman - 1981
84	A computational approach to motion perception - Uras, Girosi, et al. - 1988
83	Computations underlying the measurement of visual motion - Hildreth - 1984
82	Investigations of multigrid algorithms for the estimation of optical flow fields in image sequences - Enkelmann - 1988
79	Direction and orientation selectivity of neurons in visual area MT the macaque - Albright - 1984
73	F.: Visual Cortical Neurons as Localized Spatial Frequency Filters - Pollen, Ronner - 1983
72	Determining optical ow - Horn, Schunck - 1981
70	Parameter estimation of superimposed signals using the EM algorithm - Feder, Weinstein - 1988
64	Texture analysis using two-dimensional quadrature filters - Knutsson, Granlund - 1983
61	A model for the estimate of local image velocity by cells in the visual cortex - Grzywacz, Yuille - 1990
48	Performance of optical ow techniques - Barron, Fleet, et al. - 1992
48	Hierachical warp stereo - Quam - 1984
48	Temporal covariance model of human motion perception - SANTEN, SPERLING - 1984
42	Biological image motion processing: A review - Nakayama - 1985
41	A psychophysically motivated model for two-dimensional motion perception - Wilson, Ferrera, et al. - 1992
40	The design and use of steerable lters - Freeman, Adelson - 1991
38	Directionally selective complex cells and the computation of motion energy in cat visual cortex - Emerson, Bergen, et al. - 1992
33	Velocity estimation from image sequences with second order di erential operators - Tretiak, Pastor - 1984
32	Visual Hyperacuity : Spatiotemporal Interpolation in Human Vision - Fahle, Poggio - 1981
30	Perceived rate of movement depends on contrast - Thompson - 1982
30	A look at motion in the frequency domain - Watson, Ahumada - 1983
29	Receptive �elds, binocular interaction and functional architecture in the cat’s visual cortex - Hubel, Wiesel - 1962
26	Considerations on models of movement detection - Poggio, Reichardt - 1973
25	Human speed perception is contrast dependent - Stone, Thompson - 1992
22	Motion sensitivity and the contrast-response function of simple cells in the visual cortex - Albrecht, Geisler - 1991