Abstract: A video decoding method, which includes: decoding intra prediction related syntax elements of a current block in a bitstream, and continuing to decode a multiple reference line index of the current block in a case where multiple reference line (MRL) is allowed to be used for the current block; in a case where it is determined that an MRL index list of the current block is allowed to be sorted according to a decoding result, sorting the MRL index list of the current block, to obtain a sorted MRL index list based on the current block; and determining the reference line selected for the current block according to the multiple reference line index and the sorted MRL index list based on the current block, and performing intra prediction on the current block according to the reference line selected for the current block.

Abstract: An intra prediction fusion (IPF) method is provided. Whether a restriction condition for use of IPF for a current block is satisfied is determined when an intra prediction mode selected for the current block includes an angular mode. The use of IPF for intra prediction on the current block is restricted, when at least one of the restriction condition is satisfied.

Abstract: A video decoding method according to the present invention includes: decoding a bitstream, to obtain P quantization coefficients of a current region; determining, according to the P quantization coefficients, parity of a quantization coefficient whose parity is hidden of a first quantization coefficient; determining a target context model corresponding to the first quantization coefficient, and decoding the first quantization coefficient which is encoded based on context using the target context model, to obtain a decoded first quantization coefficient; and determining the second quantization coefficient according to the parity of the quantization coefficient whose parity is hidden and the decoded first quantization coefficient.

Abstract: A video decoder, a video encoder, and a bitstream are provided. A bitstream is decoded, to obtain P quantized coefficients and a first quantized coefficient in a current region, where the current region is a region in a current block including at least one non-zero quantized coefficient, the first quantized coefficient is obtained by performing parity hiding on a second quantized coefficient in the current region, and P is a positive integer. Then, a parity of the second quantized coefficient is determined according to the P quantized coefficients. The second quantized coefficient having the parity is determined according to the first quantized coefficient and the parity of the second quantized coefficient.

Abstract: A video encoding method is provided, which includes that: a secondary transform coefficient of a current block is obtained by decoding a code stream, the secondary transform coefficient is a transform coefficient formed by an encoding terminal performing secondary transform on a residual block of the current block; if it is determined that the use of a weighted prediction mode is allowed for an intra-frame prediction mode of the current block, a target transform kernel corresponding to the current block is selected from transform kernels of preset M types of secondary transform, M is a positive integer greater than 1; inverse secondary transform is performed on the secondary transform coefficient by using the target transform kernel to obtain a primary transform coefficient of the current block; and inverse primary transform is performed on the primary transform coefficient to obtain the residual block of the current block.

Abstract: An intra prediction method and a decoder are provided. The method includes the following. A prediction mode parameter is obtained, where the prediction mode parameter indicates that an intra prediction value is to be determined in a derivation manner. At least two prediction modes are determined according to the derivation manner. A first weight combination related to the at least two prediction modes is determined. A first cost for performing prediction on a template region of a current block in the at least two prediction modes is determined according to the at least two prediction modes and the first weight combination. A second cost for performing prediction on the template region in a first prediction mode is determined according to the first prediction mode. An intra prediction value of the current block is determined according to the first cost and the second cost.

Abstract: Embodiments of this disclosure provide a coding unit partitioning method, image coding/decoding method and apparatuses thereof. The coding unit partitioning method is by performing down-sampling on a processing unit to be partitioned to obtain a block to be partitioned of a predetermined size; inputting a first vector transformed from the block to be partitioned into a trained neural network model to acquire an output result of the trained neural network model. The output result is a partition probability in a horizontal direction and a partition probability in a vertical direction. The method includes performing a horizontal partition on the processing unit to be partitioned when the partition probability in a horizontal direction is greater than or equal to a threshold, and performing a vertical partition on the processing unit to be partitioned when the partition probability in a vertical direction is greater than or equal to the threshold.

Abstract: Embodiments of this disclosure provide an apparatus to perform a loop filter function using a convolutional neural network (CNN) and an apparatus to perform image decoding. to perform the loop filter, the apparatus is to perform down sampling on a frame of an input reconstructed image to obtain first feature maps of N channels; perform residual learning on input first feature maps of N channels among the first feature maps to obtain second feature maps of N channels; and perform up sampling on input second feature maps of N channels among the second feature maps to obtain an image of original size of the reconstructed image. Functions of the loop filter are carried out by using CNN, which may reduce a difference between a reconstructed frame and an original frame, reduce an amount of computation, and save processing time of the CNN.