A Neural Network-Based Feedforward Architecture for Recorvering 3-D Motion Information of Curved Surfaces

A neural network-based system for recovering 3-D motion information of curved surfaces from 2-D optical flow parameters is proposed in this paper. A feedforward network architecture that has explicit and concise physical meaning is adopted. A self tuning scheme with an unsupervised learning rule, that controls the dynamics of the system, is also employed. Moreover, a mechanism for preattentative focus, which effectively suppresses the spurious solution of the estimation, is embraced as well.

[1]  Shahriar Negahdaripour,et al.  Motion recovery from image sequences using only first order optical flow information , 1992, International Journal of Computer Vision.

[2]  Allen M. Waxman,et al.  Surface Structure and Three-Dimensional Motion from Image Flow Kinematics , 1985 .

[3]  Gilad Adiv,et al.  Determining Three-Dimensional Motion and Structure from Optical Flow Generated by Several Moving Objects , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Rama Chellappa,et al.  A network for motion perception , 1990, 1990 IJCNN International Joint Conference on Neural Networks.

[5]  Kenichi Kanatani,et al.  Structure and motion from optical flow under perspective projection , 1987, Comput. Vis. Graph. Image Process..

[6]  R. Chellappa,et al.  Recursive 3-D motion estimation from a monocular image sequence , 1990 .

[7]  Allen M. Waxman,et al.  Closed from solutions to image flow equations for planar surfaces in motion , 1986, Comput. Vis. Graph. Image Process..

[8]  Berthold K. P. Horn,et al.  Passive navigation , 1982, Computer Vision Graphics and Image Processing.

[9]  Jonathan A. Marshall,et al.  Self-organizing neural networks for perception of visual motion , 1990, Neural Networks.

[10]  J. I. Lipp.  Frame-to-frame image motion estimation with a fuzzy logic system , 1992, [1992] Proceedings of the 35th Midwest Symposium on Circuits and Systems.