A fast manhattan frame estimation method based on normal vectors

In most human made scenes, such as high-rise urban city or indoor environment, the surface normal vectors direction are concentrated in three orthogonal principal directions. The scene of a pattern is called Manhattan World (MW), and coordinate frame formed by directions Frame (MF). MF estimation methods have been applied to many different fields, reconstruction, Visual based Simultaneous Localization And Mapping (V-SLAM) camera calibration. this paper, we propose novel method on set vectors. A cost function axes introduced trigonometric function. For computational purpose, significantly simplified making use vector dot cross products, reduced only involves 14 scalar parameters that need be computed with O(n) complexity. experimental results show proposed has excellent real-time performance gives high accuracy both virtual real-world benchmark datasets sizes.