Using feature of nose and HAT-MARS descriptor for face recognition with pose variance
Tan Huy Thai, Thái Hoàng Lê · 2015
This paper proposes an effective method to construct a new descriptor for solving face recognition with pose variance. In this method, we use feature of nose and HAT-MARS descriptor to build a combined descriptor for matching between gallery and probe images. For experiments, our method uses only one gallery image on CMU-PIE database. The empirical results show that our method is strong and efficient.