Abstract: The mere use of supplemental enhancement information of a deep learning post-filter degrades the performance of improving image quality because filter strength fails to be adjusted in a case that features of an image vary with frame. Solution Provided is a 3D data decoding apparatus for decoding coded data into 3D data including location information and attribute information, the decoding apparatus including a header decoder configured to decode the coded data into supplemental enhancement information of a deep learning post-filter, an attribute image decoder configured to decode the coded data into an attribute image, and an attribute image filter unit configured to perform filtering processing on the attribute image, wherein the header decoder performs decoding into purpose information indicating that the attribute image filter unit performs the filtering processing on the attribute image.

Abstract: Object In the existing specifications for neural network post-filter characteristics SEI, in order to perform post-filtering processing, the SEI needs to be transmitted for each picture, which presents a problem of redundancy in a case that the same post-filtering processing is performed in all of pictures. Solution A video decoding apparatus according to an aspect of the present invention includes an image decoding apparatus configured to decode coded data and generate a decoded image, an inverse conversion information decoding apparatus configured to decode inverse conversion information as supplemental enhancement information on a per sequence basis, and a resolution inverse conversion apparatus configured to conversion the decoded image to an image with an indicated resolution by using the inverse conversion information, the resolution inverse conversion apparatus using a neural network.

Abstract: Object To reduce distortion caused by coding a mesh displacement image, and code and decode 3D data with high quality in coding and decoding 3D data using video coding scheme. Solution A 3D data decoding apparatus for decoding 3D coded data includes an image decoder configured to decode the 3D coded data into a mesh displacement image, a displacement unmapping unit configured to generate a mesh displacement from the mesh displacement image, and a coordinate system conversion unit configured to perform conversion of a coordinate system of the mesh displacement.

Abstract: A video decoding apparatus includes an image decoding apparatus configured to decode coded data to generate a decoded image, and a resolution inverse conversion apparatus configured to convert a resolution of the decoded image to a specified resolution by using inverse conversion information. The resolution inverse conversion apparatus uses a neural network. In the resolution inverse conversion apparatus, information, indicating a number of horizontal pixels and a number of vertical pixels of a patch size being a unit of processing of the neural network, is decoded. The number of horizontal pixels and the number of vertical pixels of the decoded image are set equal to a maximum value of the number of horizontal pixels and a maximum value of the number of vertical pixels of the patch size, respectively.

Abstract: Object There is the problem in the related art that the relationship between a patch size which is a unit of processing of a neural network and the sizes of input and output pictures is not clearly defined.

Abstract: According to an aspect of the present disclosure, an image decoding apparatus for decoding information specifying a neural network includes: header decoding circuitry that decodes an input tensor identification parameter specifying a process for deriving an input tensor input for a post filter of the neural network. The input tensor identification parameter is a parameter related to a color component channel.