A HEVC screen content intra-frame fast encoding algorithm based on decision tree

A technology for fast encoding and screen content, applied in the field of video encoding and decoding, it can solve problems such as reducing the complexity of video types, and achieve the effects of reducing encoding complexity, encoding time, and fast intra-frame encoding.

Active Publication Date: 2018-12-18
CHONGQING UNIV OF POSTS & TELECOMM
View PDF6 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] Although these traditional fast algorithms can effectively reduce the encoding time, most of them have limitations and canno

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 HEVC screen content intra-frame fast encoding algorithm based on decision tree
  • A HEVC screen content intra-frame fast encoding algorithm based on decision tree
  • A HEVC screen content intra-frame fast encoding algorithm based on decision tree

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0066] refer to figure 1 and figure 2 , the present invention mainly includes two modules: a CU fast division module and a PU mode selection module. The CU division includes three decision trees with depths of 0, 1, and 2. The PU mode selection includes four decision tree models with a depth of 0 to 3 and PU traversal for each layer. The model is based on the selected specific video frame, extracting the original data to form the original data set, and then generating the decision tree model through the machine learning decision tree algorithm, in which the values ​​of each branch node are provided by the machine learning algorithm, and each value is: α 01 = 16.677, β 01 =2.069,α 11 =29.5,α 12 =73.132,β 11 =3.5,β 12 =0.709,β 21 =11.5,β 22 = 51.098, gamma 01 = 11.5, gamma 11 = 21, gamma 12 =6.34,λ 11 =21,λ 21 =20.842,λ 22 = 26.53, gamma 22 = 6.131, γ 31 =20.99,λ 31 =55.05,λ 32 = 0.988.

[0067] combine image 3 In the example diagram, the image in the diagr...

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 HEVC screen content intra-frame fast encoding algorithm based on a decision tree, belonging to the technical field of video encoding and decoding. The algorithm includes: 1) extracting effective eigenvalues; 2) generating an original data set as a training set, and on the basis, training to complete a decision tree comprising a CU decision tree and a PU decision tree; 3) dividing the image into a plurality of CTUs, judging whether the current CU is a boundary CTU in the process of executing intra-frame encoding, if so, executing a standard algorithm to complete encoding, otherwise, carrying out step 4); 4) Performing fast coding, that is, completing the partition of current CU and the selection of PU mode by calculating the eigenvalue of the layer in a certain depth decision tree. The invention reduces the coding complexity by reducing the depth of CU and the mode traversal of PU through the algorithm of decision tree. Finally, it is ensured that the coding time can be effectively reduced and the effect of fast intra-frame coding can be achieved under the condition that the bit rate and peak signal-to-noise ratio are basically unchanged.

Description

technical field [0001] The present invention relates to the technical field of video encoding and decoding, in particular to a fast intra-frame encoding algorithm for HEVC screen content based on a decision tree. Background technique [0002] In recent years, with the rapid development of the Internet and mobile smart devices, video data generated by multimedia video applications accounts for about 80% of human communication data. Cisco predicts that by 2019, mobile video data will account for more than 72% of all device data communications. Although network bandwidth and storage capacity have increased rapidly in recent years, they are far from meeting people's requirements for storing and transmitting high-resolution video. In the face of such a large amount of data and the strict requirements of instant video communication, the improvement of efficient and fast video coding technology is particularly important. [0003] Video fast coding algorithm is an important techni...

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
IPC IPC(8): H04N19/96H04N19/593H04N19/105H04N19/176
CPCH04N19/105H04N19/176H04N19/593H04N19/96
Inventor 黄胜司朋涛肖傲李萌芳袁建国
Owner CHONGQING UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap