VIDEO ENCODING AND DECODING METHOD AND DEVICE

摘要

Disclosed in an embodiment of the invention are a video encoding and decoding method and device. The method comprises: acquiring a current image as well as a motion vector accuracy thereof based on the current image, the motion vector accuracy comprising a set of search step sizes, each member of the set of search step sizes being smaller than one pixel unit; acquiring a current block and a reference image thereof; determining a search start point in the reference image and setting a search step size of one pixel unit to search for a target integer pixel from the search start point; searching for a matching block for the current block from the target integer pixel, and according to the motion vector accuracy; and acquiring motion vector information and a prediction residual signal, and encoding information about the reference image, the motion vector information and the prediction residual signal into a codestream. The information about the reference image is used to indicate the reference image. The embodiment of the invention can combine video characteristics to perform encoding and decoding.