Abstract
This invention discloses the system and method that generate motion vector in order to perform the re-encoding for video signal, in which this motion vector is determined by using the sum of a basic motion vector and a micro tuning motion vector. Under the condition without using the frame-skipping technique, basic motion vector is equal to the motion vector of the present input signal and the micro tuning motion vector is determined by the magnitude of the sum of differential reconstruction error (SDRE). If the SDRE value is not greater than the predetermined threshold value, the micro tuning motion vector is zero. If the SDRE value is greater than the predetermined threshold value, the micro tuning motion vector is searched in a small region and is generated by using the optimized method of absolute error sum. Under the condition of using the frame-skipping technique, basic motion vector is generated by the sum of motion vector of the present input signal and the previously constructed motion vector of the abandoned picture. The generation method of micro tuning motion vector is the same as that of the condition without using the frame-skipping technique. The method and system disclosed in this invention can be used to improve the quality of digital video signal that needs re-encoding.