METHOD FOR PREDICTING MOTION VECTORS IN A VIDEO CODEC THAT ALLOWS MULTIPLE REFERENCING, MOTION VECTOR ENCODING/DECODING APPARATUS USING THE SAME

摘要

The present invention relates to a technology which selects an initial value when estimating a motion by using a prediction motion vector and reduces the amount of coding of a motion vector by using the prediction motion vector in a video codec which allows multiple references. A motion vector encoding equipment comprises: a prediction motion vector generating unit which generates surrounding motion vectors corresponding to a current block and generates a prediction motion vector corresponding to the current block from the generated surrounding motion vectors about each multiple reference frames; a motion estimating unit which determines a final motion vector about the current block and a final reference frame which corresponds to the final motion vector for a search area of the multiple reference frames as an object; a difference calculator which calculates a difference between the final motion vector about the current block and a final prediction motion vector which corresponds to the determined final reference frame from the prediction motion vector; and a means which encodes information about the final reference frame and the difference.