Panorama video octagon sampling method and apparatus

Abstract

The invention discloses a panorama video octagon sampling method and apparatus. The panorama video octagon sampling method includes the steps: according to a spatial definition requirement, utilizing latitude lines to divide a spherical surface into N annular curved surfaces Oi, wherein i=0, 1, ..., N-1 and N=8K, and K is an integer; utilizing longitudes to divide the annular curved surface Oi into Ti areas, wherein one of the longitudes is a 0DEG longitude; successively marking the obtained areas as Ai, j, and j=0, 1, ..., Ti-1; calculating the average value of the obtained light in every area Ai, j on the spherical surface, and taking the average value as the sampling value Si, j of a sampling point corresponding to the area; and according to the corresponding rules, arranging the sampling values into a 14K*6K matrix. The panorama video octagon sampling method and apparatus enable distribution of the sampling points to satisfy the situation that the sampling points around the equator are more and the sampling points at two poles are very few, thus avoiding redundancy of the sampling structure so as to reduce the number of the sampling points when achieving the same spatial definition.

Description

technical field [0001] The present invention relates to the technical field of video encoding and decoding, in particular to a panoramic video octagonal sampling method and device. Background technique [0002] Panoramic video refers to all the scenes around an observation point in space, which is composed of all the light that this observation point can receive. Panoramic video can be abstracted into a sphere centered on the observation point. [0003] When using a computer to process panoramic video, it is inevitable to perform discretized spatial sampling on the panoramic video. When discretizing the spatial sampling of the panoramic video, it is necessary to ensure a certain spatial sampling density to achieve the required definition. At the same time, it is also necessary to consider that the memory of the computer is not suitable for storing data of a spherical structure, and it is necessary to arrange the sampling points on the plane in a certain way. [0004] The ...

AI Technical Summary

Problems solved by technology

In the third step, when arranging, the content is often not completely adjacent, which brings disadvantages to video encoding and compression.

The distribution of sampling points of hexahedron sampling on the spherical surface is also uneven, there is still a large redundancy, and the content of the rectangle does not have complete adjacency

[0010] To sum up, the main disadvantages of the existing panoramic video sampling methods are that there is a large redundancy in the sampling structure and the resulting rectangle does not have complete adjacency of content.

The uneven distribution of sampling points on the spherical surface will not only bring sampling redundancy, but also bring inconvenience to video quality evaluation, because the spherical area corresponding to the final sampling point is different, so the distortion of each sampling point has a great impact on video quality. The impact is also different. When calculating the quality loss of the panoramic video after some processing, such as compression and decompression, it must be considered that the importance of each sampling point is different.

The non-adjacency of rectangular video in content is also not conducive to the compression coding of video coding

Images