The invention relates to the technical field of video encoding and decoding, and especially relates to a method and a device for video encoding and decoding, wherein the encoding method comprises the steps of: obtaining reconstructed and available adjacent pixel points which are adjacent to an encoded unit; finding a reference block corresponding to the encoded unit according to a motion vector of a prediction unit in the encoded unit, and obtaining reference pixel points corresponding to the reference block; calculating pixel attribute compensation parameters of the prediction unit by utilizing pixel attribute values of the adjacent pixel points and the same pixel attribute values of the reference pixel points; and calculating a first predicted value of the prediction unit according to the pixel attribute compensation parameters, and carrying out encoding according to the first predicted value. The method and the device are capable of improving the accuracy of inter-view prediction, improving encoding efficiency and avoiding the problem in the prior art that encoding pixel attribute parameters are needed since a pixel attribute parameter model is calculated by utilizing the encoded unit to be encoded.