Abstract:In the field of video content analysis and understanding, it is important to make good perception of spatial and motion features of objects in the video. However, in practice, object movement often mixes up with the camera movement which conceals the objects’ true trajectories. The camera movement, also named as global motion, which is induced by the movement of camera, is ubiquitous in the current popular we-media videos. This always makes the object trajectories related content analysis task incorrect and difficult. To reveal the true objects motion from the trajectories obtained from tracking algorithms, we propose a global motion correction procedure and obtain the real motion trajectory of the foreground object, where the global motion compensation is based on L1-norm minimization is used to solve the parametric global motion estimation. Experiments show that our algorithm could accurately estimate the inherent various global motion and restore the true motion trajectory effectively.