An Efficient Vanishing Point Detection by Clustering on the Normalized Unit Sphere

Abstract

Detecting vanishing points in digital images is important for 3D reconstruction of a scene using uncalibrated cameras. We present a new vanishing point detection algorithm based on the normalized unit sphere that not only bounds the search space but treats the finite vanishing points and the vanishing points at infinity in the same manner. The experimental results using synthetic and real data demonstrate the efficacy and performance of this algorithm.