Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Intra-frame coding optimization method and device based on convolutional neural network

A convolutional neural network and intra-frame coding technology, applied in the field of video compression, can solve problems such as instability, uncertainty, and difficulty in real-time coding, and achieve the effect of reducing the number and complexity

Active Publication Date: 2017-08-11
TSINGHUA UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] After adopting the above method, the computational complexity of the encoder can be greatly reduced in the average sense. However, there are uncertainties and instabilities in the actual encoding process, and it is difficult to complete the realization of real-time encoding.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Intra-frame coding optimization method and device based on convolutional neural network
  • Intra-frame coding optimization method and device based on convolutional neural network
  • Intra-frame coding optimization method and device based on convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0058] The technical solution of the present invention will be further described in detail in conjunction with the accompanying drawings and embodiments.

[0059] The entire intra-frame coding mode search process is to find the optimal three-combination mode from all CU / PU / TU mode combinations defined by the HEVC standard. This process is carried out according to the principle of the optimal rate-distortion cost. use vector Indicates the encoding mode of CU / PU / TU, HEVC's reference encoder takes the trouble to search all standard-defined vectors find a vector from To minimize the rate-distortion cost, the process can be expressed by the following formula:

[0060]

[0061] in, and Represented by the mode The distortion rate and bit rate for encoding, λ is the Lagrangian factor, used to adjust and In the proportion of encoding cost. In the HEVC reference code, each pattern vector must be searched to obtain the rate-distortion corresponding to the mode, and c...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates to the technical field of video compression, in particular to an intra-frame encoding optimization method and device based on a convolutional neural network. First compress the current coding unit into a data block with a size of 8×8, and calculate the edge strength of the 8×8 data block, and then determine the edge strength of the current coding unit according to the edge strength of the 8×8 data block The type is a flat coding unit, a coding unit with strong edge strength, or a coding unit with an uncertain mode. For a coding unit with an uncertain mode with a size of 32×32 or 8×8, a convolutional neural network is further used to determine the type of coding unit . By performing different coding mode processing on different classification results of the current coding unit, the number of searched coding units CU can be effectively reduced, thereby reducing coding complexity and realizing real-time coding.

Description

technical field [0001] The invention relates to the technical field of video compression, in particular to an intra-frame encoding optimization method and device based on a convolutional neural network. Background technique [0002] High Efficiency Video Coding (HEVC, High Efficiency Video Coding) is a new-generation video coding standard released by the United Video Group in 2013. Compared with the previous generation of video coding standard H.264, under the same video quality, its compression efficiency is about doubled, that is, the bit rate is reduced by about half. HEVC defines three types of units: Coding Unit (CU, Coding Unit), Prediction Unit (PU, Prediction Unit) and Transform Unit (TU, Transform Unit). For a coded frame of image, it is first divided into multiple largest coding units (LCU, Largest Coding Unit), and the LCU is divided into four sub-CUs, and each sub-CU can continue to be divided until the maximum depth. The HEVC standard defines four types of CUs...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): H04N19/593H04N19/11H04N19/14
Inventor 刘振宇余先宇汪东升
Owner TSINGHUA UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products