BAYESIAN MOTION ESTIMATION AND INTERPOLATION IN INTERLACED VIDEO SEQUENCES

Motion estimation in video is a very important tool for such applicati ons as interpolation, motion compensation, and segmentation. However, in interlaced video, it is often desirable to perform interpolation pr ior to the motion estimation. In this correspondence, a novel techniqu e is presented that, rather than minimizing the interpolation error in each of the frames used for motion estimation, seeks to make these er rors equal. The new method is shown to perform deinterlacing better th an several other approaches also based on Bayesian estimation.