A method of coding video data includes determining that chroma related syntax elements of the video data are present for luma mapping with chroma scaling (LMCS) for a coding unit (CU) of the video data, coding a syntax element of the video data when the chroma related syntax elements are present for LMCS for the CU, the syntax element is indicative of a value for determining a scaling parameter for chroma scaling in LMCS, and coding a chroma block of the CU based on the scaling parameter for chroma scaling.

A method of coding video data includes determining that chroma related syntax elements of the video data are present for luma mapping with chroma scaling (LMCS) for a coding unit (CU) of the video data, coding a syntax element of the video data when the chroma related syntax elements are present for LMCS for the CU, the syntax element is indicative of a value for determining a scaling parameter for chroma scaling in LMCS, and coding a chroma block of the CU based on the scaling parameter for chroma scaling.

A method for decoding an image signal according to the present invention may comprise the steps of: determining whether there is a brightness change between a current image including a current block and a reference image of the current image; if it is determined that there is a brightness change between the current image and the reference image, determining weight prediction parameter candidates for the current block; determining a weight prediction parameter for the current block on the basis of index information which specifies any one of the weight prediction parameter candidates; and performing a prediction on the current block on the basis of the weight prediction parameter.

A point-cloud decoding device according to the present invention includes: a tree synthesizing unit configured to decode position information on each point of point-cloud data as a decoding target, and give, in ascending order in order of decoding of the position information, an index to each point of the point-cloud data.

According to embodiments of the present document, a zero-order exponential Golomb coding scheme(ue(v)) is used for a parsing procedure of an information/syntax element related to luma/chroma ALF filter coefficient absolute values, so that operational overhead and complexity can be reduced. In addition, since the range of values of the information related to the luma/chroma ALF filter coefficient absolute values is fixed, coding that uses ue(v) can be efficiently performed.

According to embodiments of the present document, a zero-order exponential Golomb coding scheme(ue(v)) is used for a parsing procedure of an information/syntax element related to luma/chroma ALF filter coefficient absolute values, so that operational overhead and complexity can be reduced. In addition, since the range of values of the information related to the luma/chroma ALF filter coefficient absolute values is fixed, coding that uses ue(v) can be efficiently performed.

Neural image compression using substitutional quality factor learning in a latent space, including receiving a compressed bitstream and a target quality factor indicating a target compression quality, calculating a decoded latent representation of the compressed bitstream, and calculating a reconstructed image based on the decoded latent representation of the compressed bitstream and the target quality factor, computing a shared feature based on a network forward computation using shared decoding parameters (SDP) of one or more layers of a convolutional neural network, computing estimated adaptive decoding parameters (ADP) for the one or more layers of the convolutional neural network based on the shared feature, the adaptive decoding parameters, and the target quality factor, and computing an output tensor based on the estimated ADP in the one or more layers of the convolutional neural network and the shared feature.

Neural image compression using substitutional quality factor learning in a latent space, including receiving a compressed bitstream and a target quality factor indicating a target compression quality, calculating a decoded latent representation of the compressed bitstream, and calculating a reconstructed image based on the decoded latent representation of the compressed bitstream and the target quality factor, computing a shared feature based on a network forward computation using shared decoding parameters (SDP) of one or more layers of a convolutional neural network, computing estimated adaptive decoding parameters (ADP) for the one or more layers of the convolutional neural network based on the shared feature, the adaptive decoding parameters, and the target quality factor, and computing an output tensor based on the estimated ADP in the one or more layers of the convolutional neural network and the shared feature.

A method and apparatus for video processing include: determining to skip a transform process for a prediction residual based on a maximum transform size of a prediction block; and signaling the maximum transform size in a sequence parameter set (SPS).

A method and apparatus for video processing include: determining to skip a transform process for a prediction residual based on a maximum transform size of a prediction block; and signaling the maximum transform size in a sequence parameter set (SPS).