Contour Based Superquadric Tracking

This paper proposes a technique for tracking a superquadricmodelled object over a monocular video sequences. The object is currently modelled with a single superquadric. Object’s position and orientation in the first frame of the sequence are assumed known. A frame in a sequence is first processed to find object’s contour. Contour is determined by extracting edges on the frame in the vicinity of model’s contour from the previous frame. The model’s relative translation and rotation parameters are then calculated by fitting model’s contour to the frame’s contour. This fitting is achieved by minimizing the cost function, which is based on model to image mapping.