This letter presents a face normalization algorithm based on 2-D face model to recognize faces with variant postures from front-view face. A 2-D face mesh model can be extracted from faces with rotation to left or rig...This letter presents a face normalization algorithm based on 2-D face model to recognize faces with variant postures from front-view face. A 2-D face mesh model can be extracted from faces with rotation to left or right and the corresponding front-view mesh model can be estimated according to the facial symmetry. Then based on the inner relationship between the two mesh models, the normalized front-view face is formed by gray level mapping. Finally, the face recognition will be finished based on Principal Component Analysis (PCA). Experiments show that better face recognition performance is achieved in this way.展开更多
The authors found equations for complex coordinates of spectral peaks and trajectories in the case of two superposed layers, each consisting of two orthogonal gratings. The number of geometric elements in spectra was ...The authors found equations for complex coordinates of spectral peaks and trajectories in the case of two superposed layers, each consisting of two orthogonal gratings. The number of geometric elements in spectra was found for four running parameters and different number of gratings by layers. The shape of trajectories was determined in the corresponding cases. The relationships between parameters were found which could help in reducing the intervals of parameters, in particular the relationship between the inverse aspect ratios. The numerical simulation and the physical experiment were in a good agreement with the theory. The proposed technique seems to be helpful in estimation of occurrence of moir6 patterns in visual displays which makes possible the minimization in the spectral domain without calculation of spectra.展开更多
基金Supported by the National 863 Project(2001AA114140)and NNSF of China (90104013)
文摘This letter presents a face normalization algorithm based on 2-D face model to recognize faces with variant postures from front-view face. A 2-D face mesh model can be extracted from faces with rotation to left or right and the corresponding front-view mesh model can be estimated according to the facial symmetry. Then based on the inner relationship between the two mesh models, the normalized front-view face is formed by gray level mapping. Finally, the face recognition will be finished based on Principal Component Analysis (PCA). Experiments show that better face recognition performance is achieved in this way.
文摘The authors found equations for complex coordinates of spectral peaks and trajectories in the case of two superposed layers, each consisting of two orthogonal gratings. The number of geometric elements in spectra was found for four running parameters and different number of gratings by layers. The shape of trajectories was determined in the corresponding cases. The relationships between parameters were found which could help in reducing the intervals of parameters, in particular the relationship between the inverse aspect ratios. The numerical simulation and the physical experiment were in a good agreement with the theory. The proposed technique seems to be helpful in estimation of occurrence of moir6 patterns in visual displays which makes possible the minimization in the spectral domain without calculation of spectra.
