A ?;ELXAL ?;ETWORK-BASED FEEDFORW.4RD .4RCHITECTLXE FOR RECORVERING 3-11 MOTIOX LJTORMATIOK OF CURVED SURFACES
Yi Sun Mohamed, Mohamed M. Bayoumi · 1996
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 connols the dynamics of the system, is also employed. Moreo’ver, a mechanism for preattentative focus, which effectively suppresses the spurious solution of the estimation, is embraced as well.