EP-4465629-B1 - IMAGE CODING AND DECODING

EP4465629B1EP 4465629 B1EP4465629 B1EP 4465629B1EP-4465629-B1

Inventors

NISHI, TAKAHIRO
SHIBAHARA, YOUJI
SASAI, HISAO
SUGIO, TOSHIYASU

Dates

Publication Date: 20260513
Application Date: 20120112

Claims (1)

A program causing a computer to execute: an image coding method for coding a video signal using a processor, said image coding method comprising: classifying (S101) a plurality of images included in the video signal into layers, each of the plurality of images belonging to a corresponding one of the layers, the layers including a first layer and a second layer; wherein (i) when an image belonging to the first layer is coded, the image belonging to the first layer is (a) able to refer to another image belonging to the first layer and (b) is not able to refer to an image belonging to the second layer; and (ii) when an image belonging to the second layer is coded, the image belonging to the second layer is able to (a) refer to an image belonging to the first layer and (b) another image belonging to the second layer; the method further comprising coding (S102) (i) a first image (Bf16) belonging to the first layer by referring to another image belonging to the first layer and (ii) a second image (Bf8) belonging to one of the first layer and the second layer by inter prediction, wherein the second image (Bf8) precedes the first image (Bf16) in coding order; and coding (S102) a third image (Br12) belonging to the second layer by inter prediction, wherein the third image (Br12) follows the first image (Bf16) in the coding order; wherein in a case that the first image (Bf16) is a specified image to prohibit the third image (Br12) from referring to the second image (Bf8), the third image (Br12) is coded without referring to the second image (Bf8) preceding the first image (Bf16) in the coding order; and in a case that the first image (Bf16) is not the specified image to prohibit the third image (Br12) from referring to the second image (Bf8), the third image (Br12) is able to be coded by referring to the second image (Bf8) preceding the first image (Bf16) in the coding order; and an image decoding method for decoding a video signal comprising a plurality of images classified into layers, each of the plurality of images belonging to a corresponding one of the layers, the layers including a first layer and a second layer, wherein (i) when an image belonging to the first layer is decoded, the image belonging to the first layer is (i) able to refer to another image belonging to the first layer and (ii) is not able to refer to an image belonging to the second layer; and (ii) when an image belonging to the second layer is decoded, the image belonging to the second layer is able to (i) refer to an image belonging to the first layer and (ii) another image belonging to the second layer; the method further comprising decoding (S31) (i) a first image (Bf16) belonging to the first layer by referring to another image belonging to the first layer and (ii) a second image (Bf8) belonging to one of the first layer and the second layer by inter prediction, wherein the second image (Bf8) precedes the first image (Bf16) in decoding order; and decoding (S31) a third image (Br12) belonging to the second layer by inter prediction, wherein the third image (Br12) follows the first image (Bf16) in the decoding order; wherein in a case that the first image (Bf16) is a specified image to prohibit the third image (Br12) from referring to the second image (Bf8), the third image (Br12) is decoded without referring to the second image (Bf8) preceding the first image (Bf16) in the decoding order; and in a case that the first image (Bf16) is not the specified image to prohibit the third image (Br12) from referring to the second image (Bf8), the third image (Br12) is able to be decoded by referring to the second image (Bf8) preceding the first image (Bf16) in the decoding order.

Description

[Technical Field] The present invention relates to a computer program for image coding and decoding, and the like, for video. [Background Art] In an image coding method for coding video, an amount of information is generally compressed using redundancy in a spatial direction and a temporal direction held by the video. In general, conversion into a frequency region is used as the method of using the redundancy in the spatial direction, and an inter picture prediction (hereinafter referred to as inter prediction) coding is used as the method of using the redundancy in the temporal direction. When a picture is coded in the inter prediction coding, a coded picture that precedes or follows a current picture to be coded in display order is used as a reference picture. A motion vector is derived by performing motion estimation on the current picture with respect to the reference picture, and a difference is calculated between image data of the current picture and a predictive picture obtained by performing the motion compensation based on the motion vector, thereby removing the redundancy in the temporal direction. With the image coding standard called H.264 which has already been standardized, three types of pictures; that is, I-picture, B-picture, and P-picture, are used for compression of the amount of information. The I-picture is a picture on which the inter prediction coding is not performed, in other words, a picture on which intra picture prediction (hereinafter referred to as intra prediction) is performed. The P-picture is a picture on which the inter prediction coding is performed with reference to a coded picture that precedes or follows a current picture to be coded in display order. The B-picture is a picture on which the inter prediction coding is performed with reference to two coded pictures that precede or follow the current picture in display order In addition, the I-picture and the P-picture include a switching slice and the like (SI slice, SP slice) for switching between streams, and the like. With the image coding method and the image decoding method that conform to the existing H.264 standard, these methods are required to be applicable to all of the reference relations that can be conceived by the definition of the pictures of these types. When decoding a P-picture, for example, it is required that a picture that follows in the display order is allowed to be referred to. In addition, when two motion vectors are present for the B-picture, the directions of these two motion vectors may be forward or backward. In addition, pictures which are different for each block in a slice may be referred to. In order to correspond to the flexibility of such a reference structure, execution of some processes are required for the image coding method and the image decoding method in H.264 (see, for example, Non Patent Literature 1). Further aspects regarding video coding and I-pictures can be found in Non Patent Literature 2 and Patent Literature 1 and [Citation List] [Non Patent Literature] [NPL 1] ITU-TH. 264 03/2010[NPL 2] MATSUOKA S ET AL: "Coding Efficiency Improvement with Adaptive GOP Size Selection for H.264/SVC",INNOVATIVE COMPUTING INFORMATION AND CONTROL, 2008. ICICIC '08. 3RD INTERNATIONAL CONFERENCE ON, IEEE, PISCATAWAY, NJ, USA, 18 June 2008 (2008-06-18), page 356 [Patent Literature] [PL1] [WO 2007/042916 A][PL2] [EP 2 224 745 A1] [Summary of Invention] [Technical Problem] However, with the image coding method and the image decoding method disclosed by the above-described Non Patent Literature 1, there is a problem that a processing load is high. In view of the problem described above, an object of the present invention is to provide an image coding solution for allowing reducing of the processing load. [Solution to Problem] The invention is set out in the appended set of claims. In order to achieve the object described above, an image coding method according to an aspect of the present invention is an image coding method for coding a video signal, the image coding method comprising: referring to an image as a reference image for a current image to be coded, when a plurality of images included in the video signal are classified into layers such that each of the plurality of images belongs to a corresponding one of the layers, the current image being one of the plurality of images which belongs to a first layer, and the reference image belonging to a second layer located in a range restricted according to the first layer; and coding the current image based on the reference image. According to the above-described structure, the layer to which a reference image that is referred to when coding a current image to be coded belongs is present within a range that is restricted according to the first layer to which the current image belongs. Thus, since the reference image that is referred to when coding the current image is restricted according to the first layer, it is possible to reduce the processin