METHOD AND APPARATUS FOR VIDEO ENCODING, AND METHOD AND APPARATUS FOR VIDEO DECODING USING MOTION VECTOR TRACKING

摘要

A method and an apparatus for encoding and decoding an image using motion vector tracking are provided to perform predictive encoding using the more number of reference pictures, thereby improving the efficiency of prediction and encoding. A method for encoding an image using motion vector tracking comprises the following steps of: tracking the motion vector path of correspondent regions of reference pictures, to which a current block refers, and determining correspondent regions of plural reference pictures used for the prediction of the current block(1210); determining weight to apply to the correspondent regions of the plural reference pictures(1220); calculating the weights for the correspondent regions of the plural reference pictures(1230); and encoding a difference value between the current block and an estimated block to generate a bitstream(1240).