| 摘要 |
<p>Methods and systems for obtaining a motion vector between two frames of video image data are disclosed. Specifically, methods and systems of the present invention may be used to estimate a motion vector for each macroblock of a current frame with respect to a reference frame in a multi-stage operation. In a transcoder, motion estimation may be performed in the compressed domain eliminating the need to fully decode the compressed data in the first format before reencoding in a second format. In a first stage, an application implementing the process of the present invention obtains a motion vector between first and second pictures of video image data in a video sequence. Each picture includes a plurality of supermacroblocks, each having a plurality of macroblocks. For each supermacroblock of the first picture, a match supermacroblock in the second picture is determined that best matches the supermacroblock of the first picture. For each macroblock in the first picture supermacroblock, a macroblock in the second picture is determined that best matches the macroblock in the first picture supermacroblock. Macroblocks are determined by searching the second picture for a candidate macroblock that best matches the macroblock in the first picture supermacroblock. The search begins at a macroblock in the second picture supermacroblock that corresponds to the first macroblock in the first picture supermacroblock and a motion vector is determined based on the candidate macroblock. For the remaining macroblocks in the first picture supermacroblock, motion vectors are estimated based on the macroblocks in supermacroblocks of the second picture that neighbor the candidate macroblock.</p> |
