A video compression method based on deep learning

A technology of deep learning and video compression, applied in the field of video processing, to achieve the effect of reducing costs and utilizing computing power

Active Publication Date: 2022-02-15
GLOBALTOUR GROUP LTD
View PDF16 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

As a result, mobile multimedia data with video as the main body has grown rapidly, and even with the growth of high-speed optical fiber broadband and 5G technologies for mobile devices, the growth rate of video data still exceeds the development speed of technology, proposing new solutions for related video business fields. challenge

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 video compression method based on deep learning
  • A video compression method based on deep learning
  • A video compression method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] At present, in popular tourist parks, there are generally multiple amusement facilities and amusement items; each amusement item can accommodate multiple tourists at the same time for entertainment; in this regard, the park will provide tourists with photo taking, or video recording services, The purpose is to record the wonderful pictures of tourists in the process of playing, improve tourists' evaluation of the service impression of the park, and thus also hope to increase the revenue of the park; in the current photo service, the automatic shooting equipment is generally used in accordance with fixed positions or rules The route tracks and shoots the tourists, and provides the captured video within a short period of time after the tourists complete the tour, so as to maintain the enthusiasm of the tourists. Therefore, a series of equipment and systems that run the service also require high-speed response services;

[0038] Furthermore, due to the large number of touri...

Embodiment 2

[0069] This embodiment should be understood as at least including all the features of any one of the foregoing embodiments, and further improvements on the basis thereof;

[0070] During the game, the user usually plays with familiar people, such as relatives and friends, and for the recorded video images, he also hopes to keep a clear picture of the associated people at the same time; therefore, this embodiment continues to propose a method based on An implementation method for retaining clear video images of two or more users;

[0071] Wherein, when the collection module collects the facial information of the user, preferably, a first user collects the facial information simultaneously with the associated second user, so that when the collection module obtains user characteristics, it can Obtain facial feature information of more than one user; or, through the application program, the first user selects the second user associated with himself by checking the user name, conne...

Embodiment 3

[0075] This embodiment should be understood as at least including all the features of any one of the foregoing embodiments, and further improvements on the basis of it:

[0076]The amusement items in the park are relatively fixed, and their characteristics are obvious; at the same time, for the same amusement item, the settings on the photo items generally have fixed rules; Fixed, the route of video shooting and mirror movement can also have a relatively fixed track;

[0077] On the other hand, for tourists, they also hope that the video can not only leave the image information related to the characters, but also retain the information about the key events and scene characteristics at that time;

[0078] Therefore, this embodiment further optimizes this video compression method, as attached Figure 4 , including setting the second target area, so as to preserve the clarity of the non-character part of the video image; the park can propose landscapes, modeling objects, charact...

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 present invention provides a video compression method based on deep learning. The video compression method collects the user's facial information and the associated information between multiple users, and identifies and separates the video based on multiple users in the source video through deep learning. target image area, and cache the separated target image for later use; then, downsample and compress the source video, reconstruct and synthesize the compressed video based on the target image area of ​​the user, so that the source video After compression, multiple videos with clear images of the target users are obtained at the same time; this compression method takes into account both video capacity and image quality, and is conducive to the dissemination of videos among multiple target users.

Description

technical field [0001] The invention relates to the field of video processing. Specifically, it involves a deep learning-based video compression method. Background technique [0002] With the development of Internet technology and the development of video shooting technology, people are more and more fond of using video as a recording form, and sharing the video via the Internet to share their daily activities with the public. As a result, mobile multimedia data with video as the main body has grown rapidly, and even with the growth of high-speed optical fiber broadband and 5G technologies for mobile devices, the growth rate of video data still exceeds the development speed of technology, proposing new solutions for related video business fields. challenge. According to the statistics of Cisco Visual Network Index, from 2017 to 2022, global Internet traffic will increase by 3 times or more, and the peak period of traffic will increase by more than 4.8 times; and with short...

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): H04N19/70G06N3/04G06N3/08
CPCH04N19/70G06N3/08G06N3/045
Inventor 张卫平丁烨岑全李显阔
Owner GLOBALTOUR GROUP LTD
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