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

Block partitioning coding complexity optimization method and device based on deep learning method

A coding complexity and deep learning technology, applied in the field of block segmentation coding complexity optimization, can solve the problems of inability to model the CU segmentation process, limited learning ability, and shallow CNN structure.

Active Publication Date: 2018-09-04
BEIHANG UNIV
View PDF3 Cites 28 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, CNNs in the prior art have a shallow structure and limited learning ability, so they cannot accurately model the complex CU segmentation process

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
  • Block partitioning coding complexity optimization method and device based on deep learning method
  • Block partitioning coding complexity optimization method and device based on deep learning method
  • Block partitioning coding complexity optimization method and device based on deep learning method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0091] Such as Figure 6 As shown, the block segmentation coding complexity optimization method based on deep learning method of the present invention may comprise the following steps:

[0092] 101. In HEVC, check the frame coding mode currently used by the HEVC;

[0093] 102. Select a CU partition prediction model corresponding to the frame coding mode according to the frame coding mode; the CU partition prediction model is a pre-established and trained model with an early termination mechanism.

[0094] 103. Predict a CU partition result in the HEVC according to the selected CU partition prediction model, and partition the entire coding tree unit CTU according to the predicted CU partition result.

[0095] Specifically, after executing the above Figure 6 Before the shown method, the above-mentioned method may also include step 600 not shown in the following figures:

[0096] Build the ETH-CNN, train the ETH-CNN; build the ETH-LSTM, train the ETH-LSTM.

[0097] Wherein, ...

Embodiment 2

[0197] In addition, the block segmentation coding complexity optimization method based on the deep learning method in the embodiment of the present invention described above may be implemented by a block partition coding complexity optimization device. Such as Figure 7 shown.

[0198] The device for optimizing complexity of block segmentation coding based on deep learning methods may include a processor 501 and a memory 502 storing computer program instructions.

[0199] Specifically, the above-mentioned processor 501 may include a central processing unit (CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), or may be configured to implement one or more integrated circuits in the embodiments of the present invention.

[0200] Memory 502 may include mass storage for data or instructions. By way of example and not limitation, the memory 502 may include a hard disk drive (Hard Disk Drive, HDD), a floppy disk drive, a flash memory...

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 block partitioning coding complexity optimization method and device based on a deep learning method. The method comprises the following steps: checking a frame coding mode currently used by HEVC (high efficiency video coding) in the HEVC; selecting a CU partition prediction model corresponding to the frame coding mode according to the frame coding mode, wherein the CU partition prediction model is a model established and trained in advance; predicting a CU partition result in the HEVC according to the selected CU partition prediction model, and partitioning the whole CTU according to a predicted CU partition result. In the specific application, the frame coding mode is an intra-frame mode, and the CU partition prediction model is the ETH-CNN capable of being terminated in advance; if the frame coding mode is an inter-frame mode, the CU partition prediction model can the ETH-LSTM and ETH-CNN capable of being terminated in advance. Through the above method provided by the invention, the time required for decoding the CU partition in the coding is obviously shortened in the premise of guaranteeing the CU partition prediction precision, and the HEVC coding complexity is effectively reduced.

Description

technical field [0001] The present invention relates to the technical field of video coding, in particular to a block segmentation coding complexity optimization method and device based on a deep learning method. Background technique [0002] Compared with the previous generation H.264 / Advanced Video Coding (AVC) standard, the High Efficiency Video Coding (HEVC) standard can save about 50% of the bit rate under the same video quality. This benefits from some advanced video coding techniques, such as the quadtree-based coding unit (coding unit, CU) partition structure. However, these techniques also bring considerable complexity. The encoding time of state-of-the-art HEVC is about 253% longer than H.264 / AVC on average, which limits the practical application of this standard. Therefore, it is necessary to significantly reduce the complexity of HEVC coding under the premise that the rate-distortion (RD) performance is hardly affected. [0003] Over the past few years, variou...

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 Applications(China)
IPC IPC(8): H04N19/103H04N19/122H04N19/149H04N19/18H04N19/96
CPCH04N19/103H04N19/122H04N19/149H04N19/18H04N19/96H04N19/119H04N19/159H04N19/176
Inventor 徐迈李天一杨韧关振宇黄典润
Owner BEIHANG 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