3D contour image coding based on morphological filters and motion estimation

An information-lossy method for contour image sequence coding is described in this paper. It is based on motion estimation, compensation and morphological filters. In order to exploit the high temporal redundancy of the contour image sequence, motion estimation and compensation are used for the correspondent label image sequence. The first frame is directly coded by the chain-code method. All the following frames are coded in motion compensation mode. Within each frame, labels are divided into foreground labels and background labels. Only the prediction error for the foreground labels are coded. Morphological filters are applied to clean the prediction error for the foreground labels. Afterwards, an efficient prediction error coding algorithm is introduced based on the modified chain-code method. Experimental results have demonstrated outstanding performance for the coding of a contour image sequence