Adaptive global concealment of video up-conversion artefacts

Video field-rate conversion requires algorithms that improve dynamic resolution without degrading motion portrayal. Motion-compensated up-conversion is the state-of-the-art in this field. But motion compensation sometimes introduces objectionable artefacts into the output picture due to incorrect motion vectors. This paper describes a method of suppressing the image degradation due to motion compensation artefacts. It is unique in that it estimates global image degradation to determine the required level of error concealment or fallback. Two levels of fallback are implemented which take human perception properties into consideration. They are realized on a microprocessor's ROM connected to an up-converter IC.