The precision rates for small model sizes are lower than those shown in Figure 7 because the faces from the training set are unavailable for the similarity list. In this experiment ... 1 worse than it would be without the addition. P3 and P5 were anbsp;...
|Author||:||Michael S. Lew, Nicu Sebe, Chabane Djeraba, ACM Digital Library, ACM Special Interest Group on Multimedia|
|Publisher||:||Association for Computing Machinery (ACM) - 2004|