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

A GPU-based Parallel Pixel Adaptive Offset Method

Inactive Publication Date: 2019-07-09
SUN YAT SEN UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although avoiding the SAO module based on the pixel statistical classification method can obtain a large degree of time reduction, these methods are all compromised on accuracy.

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
  • A GPU-based Parallel Pixel Adaptive Offset Method
  • A GPU-based Parallel Pixel Adaptive Offset Method
  • A GPU-based Parallel Pixel Adaptive Offset Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] Such as figure 1 Shown, a kind of parallel pixel adaptive offset method based on GPU, described algorithm processes each frame of video sequence successively according to the following steps:

[0035] S1: Perform pixel-level parallel statistics on the GPU, and one thread is responsible for the statistics of one or more pixels in a certain classification mode;

[0036] S2: Relative distortion of LCU block-level parallel computing in GPU;

[0037] S3: The final SAO parameters are determined in the LCU block-level serial decision by the CPU;

[0038] S4: Perform pixel-level parallel corrections on the GPU.

[0039] In the specific implementation process, in step S1, the specific method of pixel-level parallel statistics is: one thread is responsible for the statistics of one or more pixels in a certain classification mode, and the time complexity can reach O(1). The mapping method is any one of the following three mapping methods:

[0040] (1) Aligned thread mapping me...

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 provides a parallel SAO algorithm based on a GPU. A large number of threads of the GPU are utilized for carrying out pixel level parallel statistics, LCU block level parallel relative distortion calculation and pixel level parallel correction, so that substantial time reduction is obtained. A standard SAO algorithm goes against parallelization. Firstly, a parameter fusion mechanism causes uncertain data dependency of SAO block parameters; and secondly, the regional pixel correlation leads to data synchronization waiting among a plurality of threads in the statistics process. For the above two challenges blocking SAO parallelization, two different pixel level parallel statistics algorithms based on different thread mapping method and one block level parallel relative distortion calculation algorithm are designed based on the hardware characteristics of the GPU. By a large amount of experiment analysis, compared with a serial SAO (rdSAO for short) of AVS2 standard reference software, the parallel SAO (cuSAO for short) designed and realized by the invention is substantially reduced in time consumption without precision loss, and the average time consumption reduction is as high as 70%, and more than 90% at most.

Description

technical field [0001] The present invention relates to the technical field of video coding, and more specifically, to a GPU-based parallel pixel adaptive offset method. Background technique [0002] Today, video applications have penetrated into various fields, and video coding technology is also facing more and more challenges. In April 2010, VCEG and MPEG joined forces again to form the JCT-VC working group to start formulating the new generation of video coding standard H.265 / HEVC. Soon afterwards, my country's AVS working group also began to prepare for the new generation of video coding standard AVS2 independently developed by my country. . [0003] Whether it is the international standard H.265 / HEVC or my country's AVS2 standard, the core goal is to double the compression efficiency on the basis of the previous generation of video coding standards, that is, under the premise of ensuring the same video image quality, the video stream The code rate is reduced by another...

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): H04N19/436H04N19/42
CPCH04N19/42H04N19/436
Inventor 纪庆革林润阳朱婷梁凡
Owner SUN YAT SEN 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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com