Video encoding device and video decoding device

Abstract

While maintaining a high degree of freedom in choosing partition sizes and transformation sizes adapted for local characteristics of videos, the amount of metadata is decreased. A video encoding device (10) divides an input video into blocks of a prescribed size and encodes the video block by block. The video encoding device is provided with: a prediction parameter decision unit (102) that decides the block partition structure; a predicted image generation unit (103) that generates predicted images, partition by partition, as prescribed by the partition structure; a transformation coefficient generation unit (107) which applies one of the frequency transformations included in a prescribed transformation preset to prediction residuals, i.e. the differences between predicted images and the input video; a transformation constraint derivation unit (104) which generates transformation candidate lists, i.e. lists of frequency transformations that can be applied to each partition, on the basis of partition format information; and a variable-length encoding unit (108) which, on the basis of transformation candidate lists and the transformation preset, performs variable-length encoding on transformation selection flags.

Description

technical field [0001] The present invention relates to a moving picture encoding device that encodes moving pictures to generate encoded data, and a moving picture decoding device that reproduces moving pictures based on transmitted and stored encoded data of moving pictures. Background technique [0002] <Introduction and definitions of basic terms> [0003] In the block-based video coding method, an input video to be coded is divided into predetermined processing units called macroblocks (hereinafter referred to as MB), and coding is performed for each MB to generate coded data. When reproducing a moving image, coded data to be decoded is processed and decoded in MB units to generate a decoded image. [0004] As a currently widespread block-based video coding method, there is a method (H.264 / AVC (Advanced Video Coding, Advanced Video Coding)) specified in Non-Patent Document 1. In H.264 / AVC, a prediction image estimated from an input video divided into MB units is...

AI Technical Summary

Problems solved by technology

However, a new problem arises in that the code amount of additional information required to select the partition size and transform size to be applied in decoding increases in each local area on the moving image.

However, since a large partition is likely to be selected in a region where the spatial correlation of pixel values ​​or motion vectors is high, when frequency transform with a small transform size is applied to such a partition, compared with the case of applying frequency transform with a large transform size , it is difficult to concentrate the energy of the prediction residual to a smaller number of transform coefficients

Therefore, the frequency transformation with small transform size is basically not selected, and the additional information required to select the transform size becomes wasteful

It becomes more difficult to choose a small transform size, especially if the maximum partition size is enlarged, thereby enlarging the size difference between the large partition size and the small transform size

[0026] In addition, in Non-Patent Document 2, the frequency transformation with the same size transformation size can be selected for the rectangular partition, but it does not mention what kind of criteria is used to determine the transformation that can be selected when the types of transformation sizes are further increased. size

Images