An apparatus and a method filters reconstructed images, in particular, video images, with adaptive multiplicative filters. The apparatus and method groups the multiplier coefficients of the filter into at least two groups; determines the value of each multiplier coefficient in a first group so as to be allowed to assume any value in a first set of allowed values of multiplier coefficients, determines the value of each multiplier coefficient in a second group so as to be allowed to assume any value in a second set of allowed values of multiplier coefficients, and filters the set of samples of an image with the filter. At least one of the first and second sets has at least one value that is not in the other set.

An apparatus and a method filters reconstructed images, in particular, video images, with adaptive multiplicative filters. The apparatus and method groups the multiplier coefficients of the filter into at least two groups; determines the value of each multiplier coefficient in a first group so as to be allowed to assume any value in a first set of allowed values of multiplier coefficients, determines the value of each multiplier coefficient in a second group so as to be allowed to assume any value in a second set of allowed values of multiplier coefficients, and filters the set of samples of an image with the filter. At least one of the first and second sets has at least one value that is not in the other set.

A method and apparatus for neural network based cross component prediction with scaling factors during encoding or decoding of an image frame or a video sequence, which may include training a deep neural network (DNN) cross component prediction (CCP) model with at least one or more scaling factors, wherein the at least one or more scaling factors are learned by optimizing a rate-distortion loss based on an input video sequence comprising a luma component, and reconstructing a chroma component based on the luma component using the trained DNN CCP model with the at least one or more scaling factors for chroma prediction. The trained DNN CCP may be updated for chroma prediction of the input video sequence using the one or more scaling factors, and performing chroma prediction of the input video sequence using the updated DNN CCP model with the one or more scaling factors.

A method and apparatus for neural network based cross component prediction with scaling factors during encoding or decoding of an image frame or a video sequence, which may include training a deep neural network (DNN) cross component prediction (CCP) model with at least one or more scaling factors, wherein the at least one or more scaling factors are learned by optimizing a rate-distortion loss based on an input video sequence comprising a luma component, and reconstructing a chroma component based on the luma component using the trained DNN CCP model with the at least one or more scaling factors for chroma prediction. The trained DNN CCP may be updated for chroma prediction of the input video sequence using the one or more scaling factors, and performing chroma prediction of the input video sequence using the updated DNN CCP model with the one or more scaling factors.

A method and apparatus for neural network based cross component prediction with low-bit precision during encoding or decoding of an image frame or a video sequence, which may include reconstructing a chroma component based on a received luma component using a pre-trained deep neural network (DNN) cross component prediction (CCP) model for chroma prediction, and updating a set of parameters of the pre-trained DNN CCP model with low-bit precision. The method may also include generating an updated DNN CCP model for chroma prediction with low-bit precision based on at least one video sequence, and using the updated DNN CCP model for cross component prediction of the at least one video sequence at reduced processing time.

A method and apparatus for neural network based cross component prediction with low-bit precision during encoding or decoding of an image frame or a video sequence, which may include reconstructing a chroma component based on a received luma component using a pre-trained deep neural network (DNN) cross component prediction (CCP) model for chroma prediction, and updating a set of parameters of the pre-trained DNN CCP model with low-bit precision. The method may also include generating an updated DNN CCP model for chroma prediction with low-bit precision based on at least one video sequence, and using the updated DNN CCP model for cross component prediction of the at least one video sequence at reduced processing time.

Methods of encoding and decoding depth data are disclosed. In an encoding method, depth values and occupancy data are both encoded into a depth map. The method adapts how the depth values and occupancy data are converted to map values in the depth map. For example, it may adaptively select a threshold, above or below which all values represent unoccupied pixels. By adapting how the depth and occupancy are encoded, based on analysis of the depth values, the method can enable more effective encoding and transmission of the depth data and occupancy data. The encoding method outputs metadata defining the adaptive encoding. This metadata can be used by a corresponding decoding method, to decode the map values. Also provided are an encoder and a decoder for depth data, and a corresponding bitstream, comprising a depth map and its associated metadata.

Methods of encoding and decoding depth data are disclosed. In an encoding method, depth values and occupancy data are both encoded into a depth map. The method adapts how the depth values and occupancy data are converted to map values in the depth map. For example, it may adaptively select a threshold, above or below which all values represent unoccupied pixels. By adapting how the depth and occupancy are encoded, based on analysis of the depth values, the method can enable more effective encoding and transmission of the depth data and occupancy data. The encoding method outputs metadata defining the adaptive encoding. This metadata can be used by a corresponding decoding method, to decode the map values. Also provided are an encoder and a decoder for depth data, and a corresponding bitstream, comprising a depth map and its associated metadata.

A method of processing video data including determining, for a conversion between a video and a bitstream of the video, that the bitstream includes an indicator. The indicator indicates that a first parameter set for a neural network (NN) filter model includes different filter parameters than a second parameter set for the NN filter model. The method further includes performing the conversion based on the indicator. A corresponding video coding apparatus and non-transitory computer readable medium are also disclosed.

A method of processing video data including determining, for a conversion between a video and a bitstream of the video, that the bitstream includes an indicator. The indicator indicates that a first parameter set for a neural network (NN) filter model includes different filter parameters than a second parameter set for the NN filter model. The method further includes performing the conversion based on the indicator. A corresponding video coding apparatus and non-transitory computer readable medium are also disclosed.