Finding Face in Millions, Peak of Neural Networks
Versandkostenfrei!
Versandfertig in 6-10 Tagen
27,99 €
inkl. MwSt.
PAYBACK Punkte
14 °P sammeln!
Face Recognition is one of the key research areas in computer vision. Ample work has been done on developing robust Face Recognition algorithms. Such algorithms are computationally expensive because the input face is compared / matched with all the faces present in database (also known as search space). In literature, reduced features space is preferred in order to reduce the computation. However, accuracy of reduced features space methods is quite low. The situation becomes more crucial for Large Population Face Recognition (LPFR). In this work, an effort has been made to address the aforemen...
Face Recognition is one of the key research areas in computer vision. Ample work has been done on developing robust Face Recognition algorithms. Such algorithms are computationally expensive because the input face is compared / matched with all the faces present in database (also known as search space). In literature, reduced features space is preferred in order to reduce the computation. However, accuracy of reduced features space methods is quite low. The situation becomes more crucial for Large Population Face Recognition (LPFR). In this work, an effort has been made to address the aforementioned problem. Instead of reducing the feature space the proposed technique reduces the search space. Firstly, self organizing maps (unsupervised neural networks) are used to reduce the search space in offline mode.
Andere Kunden interessierten sich für
