Patent attributes
A method for providing a stabilized digital video sequence, comprising: analyzing a set of image frames captured at different times to determine one-dimensional image frame representations; combining the one-dimensional frame representations to form a two-dimensional spatiotemporal representation of the video sequence; identifying a set of trajectories corresponding to structures in the two-dimensional spatiotemporal representation; identify a set of foreground trajectory segments and a set of background trajectory segments; analyzing the background trajectory segments to estimate a motion pattern for the digital video camera; analyzing the motion pattern for the digital video camera to determine a undesired motion portion corresponding to an unintended camera shaking motion; applying spatial shifts to at least some of the image frames of the digital video sequence to provide a stabilized digital video sequence.
