A Survey on Methods for Reconstructing Surfaces from Unorganized Point Sets Pavir i rekonstrukcijos i nestruktūrizuot ta k rinkini metod ap valga

This paper addresses the issue of reconstructing and visualizing surfaces from unorganized point sets. These can be acquired using different techniques, such as 3D-laser scanning, computerized tomography, magnetic resonance imaging and multi-camera imaging. The problem of reconstructing surfaces from their unorganized point sets is common for many diverse areas, including computer graphics, computer vision, computational geometry or reverse engineering. The paper presents three alternative methods that all use variations in complementary cones to triangulate and reconstruct the tested 3D surfaces. The article evaluates and contrasts three alternatives. Article in English iame straipsnyje apra omi ir lyginami trys metodai, skirti trima i pavir i rekonstrukcijai i nestruktūrizuot j ta k rinkini . Pavir iaus aproksimacijai naudojamas tinklelis, sudarytas i vairi dvikūgi : paprasto, nepralaid iojo ir superdvikūgio. Eksperimenti kai nustatyta, kad visi trys metodai tinka kokybi kai pavir iaus rekonstrukcijai, tik prie tai reikia sudaryti pakankamai tank pirmini ta k rinkin . Be to, nustatyta, jog rekonstrukcijos nepralaid iuoju dvikūgiu metodas tinka visi kai u dariems pavir iams, o rekonstrukcijos superdvikūgiu metodas savo sparta pranoksta kitus du metodus net 6–7 kartais. is metodas taip pat leid ia rekonstruoti pavir i ir esant itin dideliam (iki milijono) pirmini ta k rinkiniui. Straipsnis angl kalba