A method of multi-scale neural image compression with intra-prediction residuals is performed by at least one processor and includes downsampling an input image, generating a current predicted image, based on a previously-recovered predicted image, and generating a prediction residual based on a difference between the downsampled input image and the generated current predicted image. The method further includes encoding the generated prediction residual, decoding the encoded prediction residual, and generating a currently-recovered predicted image based on an addition of the current predicted image and the decoded prediction residual. The method further includes upsampling the currently-recovered predicted image, generating a scale residual based on a difference between the input image and the upsampled currently-recovered predicted image, and encoding the scale residual.

A set of rough prediction modes including a MPM subset is determined for a code block during image encoding. A first prediction mode having a mode cost less than a first threshold is selected from the set of rough prediction modes, and a second prediction mode having a mode cost less than a second threshold is selected from the MPM subset. A candidate mode subset is determined based on mode types of prediction modes contained in the set of rough prediction modes and a ranking result of mode costs of the prediction modes, when the mode cost of the first prediction mode is different from the mode cost of the second prediction mode. A target prediction mode is determined for the code block from the candidate mode subset. The code block is encoded with the target prediction mode.

A method, computer program, and computer system is provided for video encoding and decoding. Video data including one or more frames is received. One or more quantization parameters associated with the received video data are determined for frame generation or enhancement of a target frame from among the one or more frames. The video data is decoded based on the determined quantization parameters.

An image decoding method according to the present document includes: a step for deriving a luma intra prediction mode; and a step for generating a chroma candidate mode list for a chroma block including a preset number of chroma intra candidates by using at least one among a DM, a planar mode, a DC mode, a vertical mode, a horizontal mode, and an up-right diagonal mode, which are derived from the luma intra prediction mode, wherein the DM is a prediction mode corresponding to a center (CR) position of the luma block, and the chroma intra candidates other than the DM can be derived through a duplicate check of the DM and a prediction mode corresponding to any one among the top left (TL), top right (TR), bottom left (BL), and bottom right (BR) of the luma block, the planar mode, the DC mode, the vertical mode, and the horizontal mode.

An image decoding method according to the present document includes: a step for deriving a luma intra prediction mode; and a step for generating a chroma candidate mode list for a chroma block including a preset number of chroma intra candidates by using at least one among a DM, a planar mode, a DC mode, a vertical mode, a horizontal mode, and an up-right diagonal mode, which are derived from the luma intra prediction mode, wherein the DM is a prediction mode corresponding to a center (CR) position of the luma block, and the chroma intra candidates other than the DM can be derived through a duplicate check of the DM and a prediction mode corresponding to any one among the top left (TL), top right (TR), bottom left (BL), and bottom right (BR) of the luma block, the planar mode, the DC mode, the vertical mode, and the horizontal mode.

An image encoding method for encoding an image, includes: a prediction image generation step generating a synthesis-prediction image by performing synthesis processing synthesizing an inter-prediction image and an intra-prediction image to an encoding-target block; and an encoding step encoding a difference between the prediction image generated in the prediction image generation step and a pixel value of an image of the encoding-target block, a plurality of intra prediction types are allowed to be used as the intra prediction used for the synthesis processing, and the plurality of intra prediction types include an in-screen block copy prediction, the synthesis processing includes weighting processing performed to the inter-prediction image and the intra-prediction image, and a weighting parameter of the intra-prediction image in the weighting processing is determined in accordance with an intra prediction type of the encoding-target block and prediction mode combination of a plurality of blocks adjacent to the encoding-target block.

An image encoding method for encoding an image, includes: a prediction image generation step generating a synthesis-prediction image by performing synthesis processing synthesizing an inter-prediction image and an intra-prediction image to an encoding-target block; and an encoding step encoding a difference between the prediction image generated in the prediction image generation step and a pixel value of an image of the encoding-target block, a plurality of intra prediction types are allowed to be used as the intra prediction used for the synthesis processing, and the plurality of intra prediction types include an in-screen block copy prediction, the synthesis processing includes weighting processing performed to the inter-prediction image and the intra-prediction image, and a weighting parameter of the intra-prediction image in the weighting processing is determined in accordance with an intra prediction type of the encoding-target block and prediction mode combination of a plurality of blocks adjacent to the encoding-target block.

This application provides a method for decoding an unmatched pixel performed by an electronic device. The method includes: obtaining a bit stream of a target video sequence and decoding the bit stream of the target video sequence by at least two entropy decoding methods, to obtain a binary symbol string of an unmatched pixel in a target picture block, the target picture block being obtained by dividing a target picture frame in the target video sequence; inversely binarizing the binary symbol string, to obtain a component value of the unmatched pixel; and obtaining the target picture block based on the component value of the unmatched pixel. This method can improve the flexibility of coding and facilitate equalization between the coding performance and the coding overhead of the unmatched pixel.

A video compression method may include providing and transmitting encoder commands for sequencies of one or more pixels in a video frame in a YCbCr color space from a set of hierarchal encoder commands. A particular encoder command for a particular sequence may include on of a copy command, a delta command, or a make pixel command. A copy command may indicate that the particular sequence is identical to one of a previous pixel in the video frame, a pixel located above a first pixel of the particular sequence in the video frame, or a pixel in a previous video frame at a location of the first pixel. A delta command may indicate that the particular sequence includes a single pixel with color values equal to the previous pixel in the video frame combined with a signed color delta. A make pixel command may provide color values of particular sequence.

A video decoding method includes: obtaining a bitstream including a plurality of coded frames of a video signal; decoding each of the plurality of coded frames into a plurality of super blocks and each of the plurality of super blocks into a plurality of residual blocks; recovering a coded block (CB) for each of the plurality of residual blocks based on multiple reference line intra prediction (MRLP) flags and reference samples included in each coded frame, wherein multiple reference lines are divided into above-side reference lines and left-side reference lines and one above-side reference line and one left-side reference line are selected for intra prediction; reconstructing each frame of the video signal by storing the recovered CB for each of the plurality of residual blocks in a frame buffer; and continuously outputting the reconstructed frames to restore the video signal.