3D shape reconstruction using extended Kalman filter with an active camera

We propose a novel method for the 3D shape reconstruction of an object using an extended Kalman filter with a single active camera. In conventional methods, a laser scanner or stereo camera is often used as the sensor to reconstruct a 3D shape. However, they have large-scale systems and some problems are caused by them. We use only one active camera for shape reconstruction. Since an active camera takes time-series images, some points can be observed and the 3D position of the points estimated by extended Kalman filtering. Also, we consider the reconstruction of 3D geometry from the connection between each point, and plan good camera viewpoints. After estimating the 3D position of two selected points on the object and analyzing them, the active camera moves to the next viewpoint to obtain hidden information. By using these estimates of points from some planned viewpoints, 3D shape reconstruction is achieved. We apply the proposed approach to computer generated images and real world images, and we show that it is effective for shape reconstruction of an object.