A video encoding system for encoding consecutive images, the encoding of a current image being done with respect to a previous and/or subsequent image, the encoding system including a reception module to receive the current image to be encoded and to receive a non-estimated real motion vector of a moved area of the current image; a divider to divide the current image into macroblocks; a module to estimate motion vectors depending on the macroblocks of the current image and on the previous and/or subsequent image; a motion compensation module to receive motion vectors and to provide a predicted area; a module to allocate the non-estimated real motion vector to the macroblocks belonging to the moved area; a module to transmit the non-estimated real motion vector directly to the motion compensation module without any estimation of the motion vectors by the estimation module for macroblocks belonging to the moved area.