| 摘要 |
The invention relates to an image sequence coding method comprising, after a segmentation of the images into partitions with associated labels, the main steps of coding the contours of the regions sent in intra-frame mode, generating a compensated partition, and computing from said compensated partition and the current one an error partition that is then coded. While reducing the computational load, this method, that allows to transmit information about a time-varying partition with the help of the motion of the pixels, is preferably carried out by means of an encoding sub-assembly coprising an intra coding device (181), a compensating device (182), and an inter coding device (183) extracting and coding partition errors defined by an extraction of error regions where compensated and current labels are different. |
