INTERPOLATION OF MISSING DATA IN IMAGE SEQUENCES

This paper presents a number of model based interpolation schemes tail ored to the problem of interpolating missing regions in image sequence s, These missing regions may be of arbitrary size and of random, but k nown, location, The problem of locating the missing regions is discuss ed in another paper in this issue, This problem occurs regularly with archived film material, The film is abraded or obscured in patches, gi ving rise to bright and dark flashes, known as ''dirt and sparkle'' in the motion picture industry, Both 3-D autoregressive models and 3-D M arkov random fields are considered in the formulation of the different reconstruction processes. The models act along motion directions esti mated using a multiresolution block matching (BM) scheme, It is possib le to address this sort of impulsive noise suppression problem with me dian filters, and comparisons with earlier work using multilevel media n filters are performed. These comparisons demonstrate the higher reco nstruction fidelity of the new interpolators.