A method to interpret 3D motion using neural networks

This study proposes a 3D motion interpretation method which uses a neural network system consisting of three kinds of neural networks. This system estimates the solutions of 3D motion of an object by interpreting three optical flow (OF - motion vector field calculated from images) patterns of the same object obtained from three different view points. Though the interpretation system is trained using only basic 3D motions consisting of a single motion component, the system can interpret unknown multiple 3D motions consisting of several motion components. The generalization capacity of the proposed system is confirmed using diverse test patterns. Also the robustness of the system to noise is proved experimentally. The experimental results show that this method has suitable features for applying to real images.<<ETX>>

[1]  Akira Watanabe,et al.  Interpretation of Optical Flow Through Complex Neural Network , 1993, IWANN.

[2]  Thomas S. Huang,et al.  Uniqueness and Estimation of Three-Dimensional Motion Parameters of Rigid Objects with Curved Surfaces , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Jake K. Aggarwal,et al.  On the computation of motion from sequences of images-A review , 1988, Proc. IEEE.

[4]  Akira Watanabe,et al.  Interpretation of optical flow through neural network learning , 1992, [Proceedings] Singapore ICCS/ISITA `92.

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

[6]  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.

[7]  Muralidhara Subbarao.  Interpretation of Image Flow: A Spatio-Temporal Approach , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  T. Nitta.  Structure of learning in the complex numbered back-propagation network , 1994, Proceedings of 1994 IEEE International Conference on Neural Networks (ICNN'94).