Video coding and decoding method, GPU (Graphics Processing Unit) as well as interacting method and system of same and CPU (Central Processing Unit)

Abstract

The embodiment of the invention discloses an interacting method of a GPU (Graphics Processing Unit) and a CPU (Central Processing Unit). The method comprises the following steps of: reading initial data in the CPU; arraying initial data which are continuous logically and discontinuous in the physical space into data which are continuous logically and also continuous in the physical space according to preset program codes collaterally executed on the GPU; converting parallel data computed in the GPU into serial data; and transmitting the serial data onto the CPU. The embodiment of the invention also discloses an interacting system of the GPU and the CPU, the GPU and a video coding and decoding method. In the invention, the calculated load on the CPU is carried out on the GPU, and the parallel computing of the GPU is utilized to improve the video coding and decoding speed in a frame and reduce the occupation to CPU resources when broadcast quality high-definition video data are coded and decoded, so that the resources of the CPU can be released for other computations and operations in the system, and the reduction of system properties is avoided.

Description

technical field [0001] The present invention relates to the field of broadcasting and television, and more specifically, relates to a video encoding and decoding method, GPU and its interaction method and system with CPU. Background technique [0002] In recent years, the demand for high-definition video applications in the radio and television industry and individual users has gradually increased. However, due to the huge amount of high-definition video data, and the algorithm for encoding and decoding video data is also very complicated, it has led to the current mainstream platforms. When encoding and decoding, it is easy to exhaust the computing resources of the system, and the computing process is very time-consuming and inefficient, and it is difficult to meet the real-time requirements. Because the development of the CPU at this stage is limited by the architecture, technology, etc., it cannot be used at low Under the condition of low cost, the performance can be grea...

AI Technical Summary

Problems solved by technology

Although these two methods solve the above problems to a certain extent, there are inevitably some defects. The first commercial codec package is a traditional codec based on CPU, which also has the disadvantages of CPU. When encoding and decoding high-level high-definition video data, it takes up too much CPU resources, and it is easy to slow down the response of other functions in the software; the second hardware encoding and decoding card needs to use a professional hardware board, which has the following disadvantages:

[0004] 1) The cost is high. Due to the high research and development costs of professional-grade boards, the cost of purchasing such products is also high;

[0005] 2) The application flexibility is poor. Since the redundant hardware boards need to occupy one or more board slots in the PC case, a larger case and a motherboard with more slots are required, which is not flexible enough in some application environments;

[0006] 3) The grades of products are not detailed enough, resulting in a low degree of freedom for users to choose. Due to the wide range of applications in video-related fields, there are many levels from high-end professional broadcasting to low-end ordinary individual users. The company's board products often cannot meet the needs of many intermediate users;

[0007] 4) Poor scalability. When new codec standards need to be supported, users need to upgrade new boards and redeploy new versions of the software system, resulting in difficulties in extending codec standards

Images