Robust Nose Localization by Using Fitting of Ellipse from 2.5D Image

Due to the existing problem of extracting the significant feature as the nose from any viewpoint between -/+90 degree, this paper propose the new robust method for localizing the nose tip and ridge by using the ellipse fitting technique from only one 2.5D image. The cross section across the head is fitted by ellipse, and then the semimajor axis vectors are corrected to the nose tip and ridge by the nose ridge localization algorithm. Our method has the significant for several applications such as recognition, animation and computer vision. From the experimental result, the 277 samples were done. Its accuracy is 97.83%. The excellent performance of our method is to be able to localizing the nose position even though the only half of face can be captured