Spatio-Temporal Scene Level Error Concealment for Shape and Texture Data in Segmented Video Content