Video compression method and system based on super-resolution

A video compression and super-resolution technology, applied in the fields of digital video signal modification, electrical components, image communication, etc., can solve the problems of large hardware requirements, many parameters, and large number of iterations, reducing bandwidth and storage requirements, reducing The number of atoms, the effect of reducing the size

Inactive Publication Date: 2019-07-26
HANGZHOU DIANZI UNIV
View PDF7 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the super-resolution reconstruction method of JPEG compressed images based on convolutional neural network requires a deep network layer and a large amount of training, so there is a large dependence on the number of samples, a large number of parameters make training difficult, and the number of iterations required for training Large, large hardware requirements and other issues

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
  • Video compression method and system based on super-resolution
  • Video compression method and system based on super-resolution
  • Video compression method and system based on super-resolution

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] Such as figure 1 As shown, the present embodiment proposes a video compression method based on super-resolution, including:

[0054] S1. The encoder extracts high-resolution key frames and high-resolution video units based on the input high-resolution video.

[0055] Specifically, the input high-resolution video is segmented to extract high-resolution key frames. In order to make the trained sparse dictionary closer to the features of the input video, the present invention adopts the method of segmenting the video first, and then separately training an independent sparse dictionary for each sub-video. Among them, video segmentation first distinguishes video scenes, and then divides consecutive frames of the same scene into independent video units V-units.

[0056] Specifically, the shot boundary is detected from the original high-definition video in units of shots, and then high-resolution keyframes are extracted according to the shot content, which are used for subse...

Embodiment 2

[0120] Such as image 3 As shown, correspondingly, this embodiment proposes a video compression system based on super-resolution, including an encoding end and a decoding end:

[0121] The encoding end includes:

[0122] A high-resolution video processing module for extracting high-resolution key frames and high-resolution video units based on the input high-resolution video;

[0123] A low-resolution video processing module, configured to down-sample the high-resolution video to obtain a low-resolution video, and extract low-resolution key frames and low-resolution video units;

[0124] A training module, configured to train the high-resolution keyframes and low-resolution keyframes to obtain a sparse dictionary and a projection matrix;

[0125] Packing module, used to pack the sparse dictionary, projection matrix and encoded low-resolution video, and send them to the decoding end;

[0126] The decoder includes:

[0127] The code stream processing module is used to demult...

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 discloses a video compression method and system based on super-resolution. the segment-by-segment training of a mapping relation is carried out after a down-sampling video is segmented in a coding stage, and a sparse dictionary, a projection matrix and a video are transmitted together; in the decoding stage, the unique mapping matrix of each video segment is utilized to restore the low-resolution video to the original size through super-resolution, so that the requirements for bandwidth and storage are reduced, and the requirement for watching the high-resolution video is met. Onthe basis of the existing video compression coding standard, the coding code rate is greatly reduced by using the super-resolution technology through the internal information of the video, and meanwhile, the subjective quality of the decoded video is ensured as much as possible.

Description

technical field [0001] The invention belongs to the technical field of video compression, and in particular relates to a super-resolution-based video compression method and system. Background technique [0002] The information that people obtain through vision every day accounts for about 80% of the total external information. In order to meet the information requirements of modern society, video / image information often needs to be represented by digital signals, and then transmitted to the terminal through the network. With the rapid development of digital image processing and transmission technology, people's demand for high-resolution images is increasing day by day. However, the video signal has the characteristics of a large amount of information and a large bandwidth. If you want to transmit the video signal, you must solve the compression problem. It is guaranteed that the video signal can be reconstructed efficiently and high reconstruction quality can be obtained. ...

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): H04N19/126H04N19/59H04N19/132H04N19/176H04N19/192H04N19/40
CPCH04N19/126H04N19/132H04N19/176H04N19/192H04N19/40H04N19/59
Inventor 姚英彪宋晨光侯彤彤姜显扬徐欣冯维许晓荣刘晴
Owner HANGZHOU DIANZI UNIV
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