VVC intra-frame coding candidate mode rapid rate distortion cost determination method and system

Abstract

The invention discloses a rapid rate distortion cost determination method and system for a VVC intra-frame coding candidate mode, and relates to the technical field of video compression. According to the method, the rate distortion cost calculation process of the candidate mode is simplified by judging whether the candidate mode of the current intra-frame coding is the optimal intra-frame prediction mode or not. A contrast experiment result proves the effectiveness of the algorithm, and compared with a standard test platform VTM 14.0, the intra-frame coding time can be reduced by 45.78% on average on the premise of ensuring that the video coding quality is basically unchanged, and the time complexity of video coding is effectively reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of video compression, and particularly relates to a method and system for quickly determining rate-distortion cost of VVC intra coding candidate mode. BACKGROUND

[0002] With the rapid development of display technology and the increasing demand for ultra-high-definition video, video applications are developing towards high frame rate and high resolution. In order to cope with this trend, a new general video coding (VVC) coding standard is established for compressing ultra-high-definition video. Compared with the high efficiency video coding (HEVC) standard, VVC can save 50% of the coding rate under the same visual quality, but the coding complexity is also multiplied.

[0003] Intra prediction is a core part of block-based coding standards, mainly used to eliminate spatial redundancy. In the VVC coding standard, many new technologies are added to the VVC intra coding, such as multi-reference line (MRL) prediction, intra sub-partition (ISP), and intra block copy (IBC). In particular, in the VVC intra prediction coding, the number of direction prediction modes is increased from 35 in the HEVC coding standard to 67. In order to obtain the optimal one of the direction prediction modes, the rate-distortion cost of the 67 direction prediction modes usually needs to be calculated, and finally the optimal prediction mode is selected by comparing the minimum rate-distortion cost. However, this selection of the optimal mode by calculating the rate-distortion cost of all prediction modes through traversal also brings huge computational complexity to the compression coding, which seriously hinders the real-time transmission of video. Therefore, it is urgent to develop a method for quickly determining the rate-distortion cost of VVC intra coding candidate mode to solve the problem of real-time transmission of the network and improve the user experience.

[0004] Therefore, how to reduce the computational complexity of intra mode selection in the video compression process is a technical problem that needs to be solved by those skilled in the art. SUMMARY

[0005] One of the purposes of the present application is to provide a method for quickly determining the rate-distortion cost of VVC intra coding candidate mode, so as to reduce the computational complexity in mode selection.

[0006] In order to solve the above technical problems, the present application adopts the following technical scheme: a method for quickly determining the rate-distortion cost of VVC intra coding candidate mode, comprising the following steps:

[0007] S1, after the target coding block enters the rate-distortion cost calculation of the intra candidate mode, the candidate mode is judged as follows: if the current candidate mode is the first candidate mode, it is judged whether the first preset condition is met, and if the first preset condition is met, the rate-distortion cost calculation of the subsequent candidate mode is stopped; when the first preset condition is not met, it is jumped to S2;

[0008] S2, continue to calculate the rate-distortion cost of the next candidate mode, and make a judgment on the current candidate mode as follows: whether the second preset condition or the third preset condition is met, if yes, stop the rate-distortion cost calculation of the subsequent candidate mode; when the second preset condition and the third preset condition are not met, continue to jump to S2 to make a judgment on the next candidate mode until the calculation of all candidate modes of the target coding block is completed.

[0009] Preferably, the first preset condition is that the first candidate mode is a planer prediction mode.

[0010] More preferably, the expression of the second preset condition is:

[0011] RD(mode) i > RD(mode) i-1 ;

[0012] wherein RD(mode) i represents the rate-distortion cost value of the current candidate mode, and RD(mode) i-1 represents the rate-distortion cost value of the previous candidate mode.

[0013] More preferably, the expression of the third preset condition is:

[0014]

[0015] wherein Th is an adaptive adjustable threshold.

[0016] In addition, the present application also provides a system for quickly determining the rate-distortion cost of a VVC intra coding candidate mode, which comprises:

[0017] a coding block processing module for processing a target coding block;

[0018] a candidate mode judgment module for executing the above-mentioned method for quickly determining the rate-distortion cost of a VVC intra coding candidate mode.

[0019] Compared with the prior art, the present application can identify the optimal or nearly optimal candidate mode at an early stage by making preset condition judgment on the candidate mode in the calculation process, thereby avoiding complete rate-distortion cost calculation on all candidate modes. This significantly reduces the calculation amount, reduces the calculation complexity in the encoding process, and improves the encoding efficiency. Since unnecessary calculation steps are reduced, the encoding process can be completed faster. This is particularly important for real-time video transmission, which can ensure that the video content can be transmitted to the user in time and smoothly, and improve the user experience. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1A flowchart of a VVC intra coding candidate mode fast rate-distortion cost determination method in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0021] For the convenience of understanding of those skilled in the art, the present application is further described below in combination with embodiments and drawings. The content mentioned in the embodiments is not a limitation of the present application.

[0022] As shown in Figure 1 The VVC intra coding candidate mode fast rate-distortion cost determination method includes the following steps:

[0023] S1, after the target coding block enters the rate-distortion cost calculation of the intra candidate mode, the candidate mode is judged as follows: if the current candidate mode is the first candidate mode, it is judged whether the first preset condition is met, and if met, the rate-distortion cost calculation of the subsequent candidate mode is stopped; when the first preset condition is not met, it is jumped to S2;

[0024] S2, the rate-distortion cost calculation of the next candidate mode is continued, and the current candidate mode is judged as follows: whether the second preset condition or the third preset condition is met, and if met, the rate-distortion cost calculation of the subsequent candidate mode is stopped; when the second preset condition and the third preset condition are not met, it is continued to jump to S2 to judge the next candidate mode until the calculation of all candidate modes of the target coding block is completed.

[0025] In the above steps, the first preset condition is that the first candidate mode is a planer prediction mode. The expression of the second preset condition is:

[0026] RD(mode) i >RD(mode) i-1 ;

[0027] Wherein RD(mode) i represents the rate-distortion cost value of the current candidate mode, and RD(mode) i-1 represents the rate-distortion cost value of the previous candidate mode.

[0028] The expression of the third preset condition is:

[0029]

[0030] Wherein Th is an adaptive adjustable threshold.

[0031] The fast determination method disclosed in the embodiment is executed on a test platform of video compression reference software VTM14.0, on an Inter(R)Core(TM)i7 CPU, 3GB RAM PC, to evaluate the feasibility and effectiveness of the method. The general test video sequences of different resolutions and different texture contents include “BasketballDrive”, “BQTerrace”, “Cactus”, “MarketPlace”, “RitualDance”, “BasketballDrill”, “BQMall”, “PartyScene”, “RaceHorsesC”, “BasketballPass”, “BlowingBubbles”, “BQSquare”, “RaceHorses”, “FourPeople”, “Johnny”, “KristenAndSara”. The coding configuration is random access (All Intra, AI), and four coding quality parameter (quality parameters, QP) combinations are used, that is, 22, 27, 32 and 37, for compression test under different code rates. For video compression performance evaluation, the changes of coding code rate and coding quality (BD-rate) are used. The coding time saving is defined as:

[0032]

[0033] In the formula, i represents four different QP values, T Ori is the compression coding time of the original test model, T Pro is the compression coding time after the application of the method of the application to the original test model.

[0034] Referring to Table 1, Table 1 is the performance comparison result of the method provided in the embodiment on the VTM14.0 test platform, wherein Th=10.

[0035] Table 1. Performance comparison result of the method provided in the embodiment on the VTM14.0 test platform (%)

[0036]

[0037]

[0038] From the comparison results of BD-rate and coding time saving in Table 1, it can be seen that the method provided in the embodiment can save an average of 45.78% of coding time, and the average increase of BD-rate is only 2.38%, so the method of the embodiment effectively reduces the VVC intra coding time.

[0039] In summary, the fast determination method provided by the present application can greatly reduce the calculation complexity of the intra prediction mode while hardly reducing the video compression quality. Meanwhile, the method is simple in design, does not refer to any space-time domain related information, and is very easy to integrate into a parallel compression encoding framework to further reduce the VVC intra compression encoding time.

[0040] In order for those skilled in the art to more conveniently understand the improvements of the present application over the prior art, some of the drawings and descriptions of the present application have been simplified, and the above embodiments are the preferred implementation of the present application, in addition to which, the present application can also be implemented in other ways, and any obvious substitutions within the concept of the present technical solution are within the protection scope of the present application.

Claims

1. A method for fast rate-distortion cost determination of VVC intra coding candidate modes, characterized in that, The method comprises the following steps: S1, after the rate-distortion cost calculation of the target coding block entering the intra candidate mode, the candidate mode is judged as follows: if the current candidate mode is the first candidate mode, it is judged whether the first preset condition is met, and if met, the rate-distortion cost calculation of the subsequent candidate mode is stopped; when the first preset condition is not met, S2 is jumped to; S2, the rate-distortion cost calculation of the next candidate mode is continued, and the current candidate mode is judged as follows: whether the second preset condition or the third preset condition is met is judged, and if met, the rate-distortion cost calculation of the subsequent candidate mode is stopped; when the second preset condition and the third preset condition are not met, S2 is continued to jump to, the next candidate mode is judged, and the calculation of all candidate modes of the target coding block is completed.

2. The method of Claim 1, wherein, The first preset condition is that the first candidate mode is a planer prediction mode.

3. The method according to claim 2, characterized in that: The expression of the second preset condition is: RD(mode) i >RD(mode) i-1 ; Wherein RD(mode) i RD(mode) i-1 RD(mode) 4. The method according to claim 3, characterized in that: The expression of the third preset condition is: Wherein Th is an adaptive adjustable threshold.

5. A system for fast rate-distortion cost determination of VVC intra coding candidate modes, characterized in that, Comprise: A coding block processing module for processing a target coding block; A candidate mode judgment module for executing the method for determining the rate-distortion cost of the VVC intra coding candidate mode quickly according to any one of claims 1 to 4.

Images