Model-based 3-D Shape and Motion Estimation Using Sliding Textures

Given an accurate 3-D shape model of a scene, the motion parameters of a moving camera can be recovered with high accuracy using model-based motion estimation techniques. Shape errors, however, reduce the accuracy of this kind of motion estimation considerably. In this paper, model-based motion estimation is combined with simultaneous object shape refinement. A deformable 3-D shape model of low dimensionality is employed to approximate the object shape. Camera position and orientation for all views as well as object shape refinements are estimated simultaneously from the image data using an optical-flow-based approach. In comparison to traditional flexible body motion estimation, our formulation of the shape deformation allows the object texture to slide on the object surface. Experimental results illustrate that combined shape and motion estimation using sliding textures improves the calibration data of the individual views in comparison to fixed-shape modelbased camera motion estimation.