Single-sample face recognition based on mirrows and singular value decomposition
Mei Wang · Microcomputer & its Applications · 2010
Nowdays,there are many methods to deal with face recognition,when there are a sufficient number of training samples these methods can achieve good recognition results,but to deal with a single sample face recognition problem,the effects decreased rapidly.To overcome this problem,a new method which is called image-based singular value decomposition is presented,through the use of mirrors to increase the imformation of training samples.Experiments show that the proposed approach is better than the others,and to some extent,it can also overcome the effect of changes in attitude.