VIDEO ENCODING AND DECODING WITH INCREASED ERROR RESISTANCE

摘要

1. A method of encoding a sequence of digital images into a bitstream, wherein at least one image portion is encoded by motion compensation relative to a reference image portion, and for at least one encoded image portion, the method comprises generating a set of motion information predictors for the encoded portion images using the target number of motion information predictors, the generation step comprising sub-steps in which: receive a first set of motion information predictors, each of which is associated with a portion of an image having a predetermined spatial and / or temporal relationship with said encoded image portion, modifying said first set of motion information predictors by removing one or more duplicated motion information predictors to obtain a reduced set of predictors motion information containing the first number of motion information predictors, after receiving the mentioned of the abbreviated set of motion information predictors, compare said first number of motion information predictors with said target amount, and if said first quantity is less than said target amount, an additional motion information predictor is added and said additional motion information predictor is added to said reduced motion information predictor set. 2. The method of claim 1, wherein the predictors of motion information from said first set of explicit