Robust computation of optical flow in a multi-scale differential framework

The authors developed an algorithm for computing optical flow in a differential framework. The image sequence is first convolved with a set of linear, separable spatiotemporal filters similar to those that have been used in other early vision. The brightness constancy constraint can then be applied to each of the resulting images, giving in general, an overdetermined system of equations for the optical flow at each pixel. There are three principal sources of error: (a) stochastic due to sensor noise, (b) systematic errors in the presence of large displacements, and (c) errors due to failure of the brightness constancy model. The analysis of these errors leads to development of an algorithm based on a robust version of total least squares. Each optical flow vector computed has an associated reliability measure which can be used in subsequent processing. The performance of the algorithm on the data set used by J. Barron et al. (1992) compares favorably with other techniques. In addition to being separable, the filters used are also causal, incorporating only past time frames. The algorithm is fully parallel and has been implemented on a multipleprocessor machine.<<ETX>>

[1]  Sabine Van Huffel,et al.  Total least squares problem - computational aspects and analysis , 1991, Frontiers in applied mathematics.

[2]  Takeo Kanade,et al.  An Iterative Image Registration Technique with an Application to Stereo Vision , 1981, IJCAI.

[3]  Karl Pearson F.R.S..  LIII. On lines and planes of closest fit to systems of points in space , 1901 .

[4]  Berthold K. P. Horn,et al.  Determining Optical Flow , 1981, Other Conferences.

[5]  Jitendra Malik,et al.  Determining Three-Dimensional Shape from Orientation and Spatial Frequency Disparities , 1991, ECCV.

[6]  R. Young.  GAUSSIAN DERIVATIVE THEORY OF SPATIAL VISION: ANALYSIS OF CORTICAL CELL RECEPTIVE FIELD LINE-WEIGHTING PROFILES. , 1985 .

[7]  Kenji Mase,et al.  Simultaneous multiple optical flow estimation , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

[8]  P. Sprent.  Models in regression and related topics , 1971 .

[9]  David J. Heeger,et al.  Optical flow from spatialtemporal filters , 1987 .

[10]  P Perona,et al.  Preattentive texture discrimination with early vision mechanisms , 1990 .

[11]  B Julesz,et al.  Cooperative and non-cooperative processes of apparent movement of random-dot cinematograms. , 1985, Spatial vision.

[12]  A. Verri,et al.  Differential techniques for optical flow , 1990 .

[13]  John F. Canny,et al.  A Computational Approach to Edge Detection , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[14]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[15]  Alessandro Verri,et al.  Computing optical flow from an overconstrained system of linear algebraic equations , 1990, [1990] Proceedings Third International Conference on Computer Vision.

[16]  Vishal Markandey,et al.  Total least squares fitting spatiotemporal derivatives to smooth optical flow fields , 1992, Defense, Security, and Sensing.

[17]  Michael J. Black,et al.  A framework for the robust estimation of optical flow , 1993, 1993 (4th) International Conference on Computer Vision.

[18]  E. Adelson,et al.  Early vision and texture perception , 1988, Nature.

[19]  Edward H. Adelson,et al.  Probability distributions of optical flow , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[20]  Jitendra Malik,et al.  Computational framework for determining stereo correspondence from a set of linear spatial filters , 1992, Image Vis. Comput..

[21]  Hans-Hellmut Nagel,et al.  On the Estimation of Optical Flow: Relations between Different Approaches and Some New Results , 1987, Artif. Intell..

[22]  Claude L. Fennema,et al.  Velocity determination in scenes containing several moving objects , 1979 .

[23]  A. Madansky.  The fitting of straight lines when both variables are subject to error , 1959 .

[24]  O. Braddick,et al.  Direction discrimination for band-pass filtered random dot kinematograms , 1990, Vision Research.