Multi-view video code rate control method based on key frame code rate-quantitative model

Abstract

The invention discloses a multi-view video code rate control method based on a key frame code rate-quantitative model. Code rate control is carried out at four layers including a view layer, a GOP layer, a frame layer and a macro block layer, and code rates among view points are allocated according to scale values of actual coding of the view points in the view layer; the total code rate of the whole GOP is allocated in the GOP layer and a coding quantization parameter of the key frame of each GOP is calculated according to the key frame code rate-quantitative model of the key frame, wherein the key frame code rate-quantitative model is obtained through analysis; code rates are allocated in the frame layer according to code rate allocation weights of B frame of different levels reasonably; and code rates of macro blocks are allocated in the macro block layer according to coding complexities of the macro blocks and coding quantization parameters of the macro blocks are solved. The multi-view video code rate control method has the advantages that the code rate-quantization model of the key frame is utilized, and an unequal code rate allocation strategy is adopted for the B frames of different levels, so that target bit allocation can be carried out more reasonably, code rate control is effectively carried out, and better rate-distortion performance and subjective quality are provided under the condition that code rate control precision is ensured.

Description

technical field [0001] The invention relates to a multi-viewpoint video code rate control technology, in particular to a multi-viewpoint video code rate control method based on a key frame code rate-quantization model. Background technique [0002] In order to meet people's increasingly high diversity requirements for video quality and content, multi-view video technology with 3D vision function is attracting more and more attention from academia and industry, and has become one of the hot spots of video research in recent years. . Bit rate control is one of the key technologies in multi-view video related applications. It needs to allocate bit rate reasonably in time to prevent buffer overflow. The balance of video quality between. The international standardization organization JVT (Joint Video Team) has been working on the formulation of multi-view video coding standards, and proposed JMVC (Joint Multi-view Video Coding) as a public research platform. In JMVC, the HBP (...

AI Technical Summary

Problems solved by technology

In the traditional single-channel rate control algorithm, all quantization parameter adjustments are only for P frames or B frames, and I frames are only encoded using the average QP (QuantizationParameter) value of all P frames / B frames in the previous GOP , if the average QP value of all P frames / B frames of the previous GOP is large, it will lead to a large QP selection of the I frame, which will reduce the image quality of the entire sequence

Images