Joint Detection, Interpolation, Motion and Parameter Estimation for Image Sequences with Missing Data

This paper presents methods for detection and reconstruction of ‘missing’ data in image sequences which can be modelled using 3-dimensional autoregressive (3DAR) models. The interpolation of missing data i s i m portant in m a n y areas of image processing, including the restoration of degraded motion pictures, reconstruction of drop-outs in digital video and automat ic ‘re-touching ’ of old photographs. Here a probabilistic Bayesian framework i s adopted. T h e method assumes n o pr ior knowledge of the motion field or 3DAR model parameters as these are estimated jointly with the missing image pixels. Incorporating a degradation model in to the framework allows detection t o proceed jointly with interpolation.