Abstract: This application relates to the field of artificial intelligence technologies, and provides a feature map processing method and a related device. The method is implemented by invoking a neural network model that includes a plurality of input adaptation branches and a post-processing part. An output of each of the plurality of input adaptation branches is an input of the post-processing part, and each of the plurality of input adaptation branches can downsample a feature map at a different ratio.

Abstract: This application provides picture or audio encoding and decoding methods and apparatuses, and relates to the field of artificial intelligence (AI)—based picture or audio encoding and decoding technologies, and specifically, to the field of neural network-based picture feature map or audio feature variable encoding and decoding technologies. The encoding method includes: obtaining a to-be-encoded target, where the to-be-encoded target includes a plurality of feature elements, and the plurality of feature elements include a first feature element. The method further includes: obtaining a probability estimation result of the first feature element; determining, based on the probability estimation result of the first feature element, whether to perform entropy encoding on the first feature element; and performing entropy encoding on the first feature element only when it is determined that entropy encoding needs to be performed on the first feature element.

Abstract: This application relates to the field of artificial intelligence (AI)-based video or picture compression technologies. The decoding method includes: performing probability estimation on input data by using a neural network, to obtain probability distribution of residual values of a plurality of samples and included in a residual of a current picture, where the input data includes at least a residual of a reference picture, and the reference picture is a decoded picture obtained before the current picture is decoded; performing arithmetic entropy decoding on a bitstream based on the probability distribution of the residual values of the plurality of samples and included in the residual of the current picture, to obtain first entropy decoding data, where the first entropy decoding data represents the residual of the current picture; and obtaining a reconstructed sample value of the current picture based on the residual of the current picture.

Abstract: This application provides a scalable encoding and decoding method and apparatus. The encoding method includes: applying a control signal of a first signal component of a video signal to a first feature map of the first signal component, to obtain a second feature map of the first signal component, where the control signal of the first signal component is obtained through learning; applying a control signal of a second signal component of the video signal to a first feature map of the second signal component, to obtain a second feature map of the second signal component, where the control signal of the second signal component is obtained through learning; and obtaining a bitstream of the video signal based on the second feature map of the first signal component and the second feature map of the second signal component.

Abstract: The technology of this application relates to an entropy encoding method that includes obtaining base layer information of a to-be-encoded picture block, where the base layer information corresponds to M samples in the picture block, and M is a positive integer, obtaining K elements corresponding to enhancement layer information of the picture block, where the enhancement layer information corresponds to N samples in the picture block, both K and N are positive integers, and N?M, inputting the base layer information into a neural network to obtain K groups of probability values, where the K groups of probability values correspond to the K elements, and any group of probability values is for representing probabilities of a plurality of candidate values of a corresponding element, and performing entropy encoding on the K elements based on the K groups of probability values.

Abstract: A motion vector prediction method and device, includes determining a motion vector type of the motion vector of the current block based on a reference frame type of the reference frame, obtaining a motion vector type of a first motion vector of at least one spatially adjacent prediction block of the current block when the at least one spatially adjacent prediction block is an inter-frame prediction block, and when the obtained motion vector type of the first motion vector is different from all motion vector types of motion vectors of the current block, determining a second motion vector of a spatially adjacent pixel block of the current block based on decoding information of the spatially adjacent pixel block, and using the second motion vector or a scaling value of the second motion vector as a candidate motion vector predictor of the current block.

Abstract: A motion vector prediction method and device, includes determining a motion vector type of the motion vector of the current block based on a reference frame type of the reference frame, obtaining a motion vector type of a first motion vector of at least one spatially adjacent prediction block of the current block when the at least one spatially adjacent prediction block is an inter-frame prediction block, and when the obtained motion vector type of the first motion vector is different from all motion vector types of motion vectors of the current block, determining a second motion vector of a spatially adjacent pixel block of the current block based on decoding information of the spatially adjacent pixel block, and using the second motion vector or a scaling value of the second motion vector as a candidate motion vector predictor of the current block.