| 摘要 |
The invention provides a method and apparatus for determining a motion vector between a current frame and its reference frame of video signals. A search block is motion-estimated with respect to its corresponding search region to thereby select a predetermined number of candidate blocks among the candidate blocks included in the corresponding search region, wherein the selected candidate blocks have error functions not larger than those of the unselected candidate blocks. Thereafter, error signals for each of the selected candidate blocks are obtained and transformed into a set of transform coefficients. Next, an optimum error signal is selected based on the sets of transform coefficients and, a displacement of pixels between the search block and a candidate block which corresponds to the optimum error signal is selected as a motion vector for the search block. |
