| 摘要 |
The present invention relates to a system and method for automatic extraction of a 3D digital image object based on a 2D digital image. According to the present invention, a plurality of feature points are extracted from a 2D digital image by using a Voronoi tessellation algorithm. Clusters for the respective extracted feature points are constructed and then the individual clusters are implemented in 3D by using depth information of the 2D digital image, thereby enabling the 2D digital image to be automatically converted into a 3D digital image object. Accordingly, according to the present invention, there are advantages in which a 3D digital image object can be automatically extracted from a 2D digital image at a high speed, and a 3D digital image at a time desired by a user can be acquired through operation of camera walking. The system comprises: a 2D digital image storage module which stores a 2D digital image; a 3D digital image object conversion module which converts the 2D digital image into a 3D digital image object; a frame storage module which stores the 3D digital image object; and a display module which displays the stored 3D digital image object or 3D digital image object, on which camera direction has been performed by a camera walk module, on a user terminal. |
