Title :
Artificial neural network system for 3-D motion perception
Author :
Sun, Yi ; Bayoumi, Mohamed M.
Author_Institution :
Dept. of Electr. Eng., Queen´´s Univ., Kingston, Ont., Canada
Abstract :
This paper proposes an artificial neural network system that estimates the 3-D motion and structure parameters of curved surfaces from measured 2-D optical flow parameters. The system is constructed based on the assumption that the optical flow measurement is available and that the object in the scene can be approximated by patches of curved surfaces.
Keywords :
image sequences; motion estimation; neural nets; parameter estimation; 3D motion perception; artificial neural network system; curved surfaces; measured 2D optical flow parameters; motion parameter estimation; structure parameter estimation; Artificial neural networks; Fluid flow measurement; Image motion analysis; Motion measurement; Neural networks; Nonlinear optics; Optical computing; Optical fiber networks; Optical imaging; Particle beam optics;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.716775
