Face recognition with non-negative matrix factorization
Menaka Rajapakse, Lonce L. Wyse · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2003
A face can conceptually be represented as a collection of sparsely distributed parts: eyes, nose, mouth etc.We use Non-negative Matrix Factorization (NMF) to yield sparse representation of localized features to represent distributed parts over a human face. This paper explores the potential of NMF for face recognition and the possibilities for gender-based features in face reconstruction. Further, we compare the results of NMF with other common face recognition methods.