Relative posture estimation using high frequency markers

Recently, research fields of augmented reality and robot navigation are actively investigated. Estimating a relative posture between an object and a camera is an important task in these fields Visual markers are frequently used to estimate a relative posture between an object and a camera, but the usage of visual markers spoils a scene. In this paper, we propose a novel method for posture estimation by using kernel regressions and high frequency markers that do not spoil a scene. The markers are embedded in an object´s texture in the high frequency domain and it is hard for people to notice the markers. We observe changes of spatial frequency of object´s texture by using a high-definition camera to estimate the current posture of an object. We show the effectiveness of our method by experimental results with both simulated and real data.