| 摘要 |
<P>PROBLEM TO BE SOLVED: To provide a method for converting a moving image code string by which the moving image code string is re-encoded in accordance with the application, and the throughput can be reduced at the time of performing the re-encoding without lowering the encoding efficiency by predicting a motion vector used for predicting motion compensation. <P>SOLUTION: In this method, an original motion vector canceled in the conventional example is held. When first and second inter-frame distances FD1 and FD2 are different from each other, the first motion vector MV1 of a block in a new objective image frame FO1 with respect to a block in a first image frame FR1 is calculated by utilizing not only the second motion vector MV2 of a block in an original image frame FO2 with respect to a block in a second image frame FR2, but also another original motion vector (for example, the fourth motion vector MV4 of a block in a third image frame FR3 with respect to a block in a fourth image frame FR4 or the fifth motion vector MV5 of a block in the second image frame MV2 with respect to a block in a fifth image frame FR5). <P>COPYRIGHT: (C)2003,JPO |
