A video processing method includes determining an image before a current frame in time sequence as a long-term reference image according to an identifier, determining a motion vector candidate list of a current coding unit, receiving a plurality of indexes, and obtaining an initial motion vector of the current coding unit from the motion vector candidate list according to the indexes. If a reference image pointed to by the initial motion vector is a short-term reference image, the initial motion vector is corrected to obtain a corrected motion vector and prediction is performed for the current coding unit according to the corrected motion vector. If the reference image is a long-term reference image, prediction is performed for the current coding unit according to the initial motion vector without correcting the initial motion vector.