Method and device for implementing interframe forecast based on relativity between future positions

Abstract

The invention relates to an intra-frame prediction method and an intra-frame prediction device based on a correlation between prediction points in the video technical field. Firstly, the invention receives a macro block type, a prediction model parameter, an upside reference pixel and a left reference pixel. Then according to the received parameters of the prediction model and the macro block type, the upside reference pixel and the left reference pixel, a prediction computation of the current macro block and a renewal of the reference pixel are completed. Finally a computed prediction value is sent to reconfigurable modules outside. The device of the invention comprises a reference pixel register set, a reference pixel control unit, a basic operation unit and an output control unit. The invention supports H. 264 high-end specification and AVS standards to improve comprehensive frequency of design and at the same time has the advantages of low cost for hardware resource and low power consumption under the same data processing capability.

Description

technical field [0001] The present invention relates to a method and device in the technical field of telecommunications, in particular to a method and device for implementing intra-frame prediction based on correlation between prediction points. Background technique [0002] Video technology is widely used in the transmission and storage of video data, from which the original image can be fully or approximately restored. The intra-frame prediction technology uses the spatial correlation of image data to predict all current pixels using the nearest decoded pixel values ​​on the left and upper sides of the current block. In the H.264 standard, the smallest block processing unit defined is a 4x4 block, with 9 prediction modes for luma and 4 prediction modes for chroma; in the AVS standard, the smallest block processing unit defined is an 8x8 block , there are 5 prediction modes for luma, and 4 prediction modes for chroma. Among the prediction modes adopted by the AVS standar...

AI Technical Summary

Problems solved by technology

But there are following disadvantages: 1. Due to the adoption of a large amount of control logic, it can only work at a lower clock frequency (55MHz), thus failing to meet the requirements for real-time processing of high-definition video data (for 1920x1080, 4:2:0 format video, processing 30 frames per second); 2. Due to the use of 4-way parallel processing architecture, the hardware overhead is still relatively large

Images