Few images based 3D Reconstruction of Korean Wooden Architecture Bracketing System(Gongfo)

Jong-Gook Ko, Seungjae Lee

Abstract: Three-dimensional (3D) reconstruction from 2D input images is a fundamental task in computer vision and has numerous applications in fields such as robotics, augmented reality, and cultural heritage preservation. This paper presents a technique for performing 3D reconstruction of Korean Wooden Architecture Bracketing using Structure-from-Motion(SfM) and Multi-View_Stereo(MVS) technology. SfM leverages the geometric relationships between multiple images to estimate camera poses and reconstruct the underlying 3D scene. the MVS algorithm is used to complete the dense reconstruction of objects. We show that few iages based 3D reconstrution results for korean wooden architecture bracketing system named by Gongfo in korean.

External IDs:dblp:conf/ictc/KoL23