METHOD FOR ENCODING AND DECODING A MULTI-VIEW VIDEO AND APPARATUS THEREFOR

摘要

A method for encoding and decoding multi-view images and an apparatus for the same are provided to reduce information amount of a motion vector differential value and a disparity differential value and to improve the prediction encoding efficiency of the apparatus by using a prediction mode judging part and a predictor generating part. An apparatus for calculating an image information predictor in multi-view image video coding comprises a prediction mode judging part(110), and a predictor generating part(350). The prediction mode judging part discriminates the prediction mode of the present frame on the basis of the direction and the number of a reference frame. The predictor generating part calculates the predictor of the image information of the present block. The predictor generating part has a motion vector prediction part(120), and a disparity prediction part(30). The motion vector prediction part calculates a motion vector predictor of the present block. The disparity prediction part calculates the disparity predictor of the present block.