Method and device for error concealment in motion estimation of video data

An encoding method of encoding a first frame and a second frame of a video bit stream, each frame of the video bit stream being defined by a plurality of blocks of pixels, the method comprising:

• extracting a first motion vector field containing a plurality of motion vectors of the first frame, the first motion vector field consisting of motion vectors relating to a plurality of macroblocks of the first frame, which macroblocks each contain a plurality of blocks, each block having an associated motion vector;
• extracting a second motion vector field containing a plurality of motion vectors of the second frame that precedes or follows the first frame, the second motion vector field consisting of a motion vectors relating to a plurality of macroblocks of the second frame, which macroblocks each contain a plurality of blocks, each block having an associated motion vector;
• determining an estimated motion vector field of the first frame by projecting the extracted second motion vector field of the second frame to the first frame;
• generating at least one transformation parameter that represents a single motion transformation that when applied to each of the plurality of motion vectors of the estimated motion vector field generates a motion vector field that is similar to or the same as the first motion vector field of the first frame;
• and
• transmitting the generated transformation parameter and the motion vector field of the first frame to a decoder wherein the generated parameter can be used by the decoder, if a part of the first frame is not received by the decoder, to obtain an enhanced estimate of a motion vector of the first frame relating to the part of the frame that is not received by the decoder by applying the single motion transformation to the estimated motion vector field of the first frame.