Unlock instant, AI-driven research and patent intelligence for your innovation.

A Deep Learning-Based Restoration and Reproduction Method for Old Movies

A deep learning, old technology, applied in the field of deep learning and computer vision, to achieve the effect of high image clarity, high stability, and reduced consumption

Active Publication Date: 2022-02-15
福建帝视科技集团有限公司
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

If you want to finely repair frame by frame, it will take at least a few months for a movie, and the cost is also in the millions

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 Deep Learning-Based Restoration and Reproduction Method for Old Movies
  • A Deep Learning-Based Restoration and Reproduction Method for Old Movies
  • A Deep Learning-Based Restoration and Reproduction Method for Old Movies

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0081] Such as Figure 1-3 As shown in one of the above, the present invention proposes a method for repairing and reproducing old movies based on deep learning. The repair process mainly includes de-interlacing, video de-noising, video de-blurring, video interpolation and super-resolution technology. The specific process is as figure 1 shown. All convolution layers in the present invention use a convolution kernel with a size of 3×3, and the specific steps are as follows:

[0082] Step 1: Extract frames from the video through ffmpeg, and form the training data set of the deinterlacing model, the training data set of the video frame interpolation model, the training data set of the deblurring network and the training data set of the super-resolution model;

[0083] Step 1.1: Form a training data set for the deinterlacing model (model1):

[0084] Step 1.1.1: Extract frames from the video through ffmpeg to obtain each frame of image, scan the obtained video frames in even fie...

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 method for repairing and reproducing old videos based on deep learning. The training data set of the resolution model; Step 2: Train the deinterlacing network model; Step 3: Train the video frame insertion network model; Step 4: Train the deblurring network; Step 5: Train the super-resolution network; Step 6: Train the noise network. Based on deep learning, the invention applies de-interlacing, video denoising, video deblurring, video frame insertion and super-resolution technology to repair old videos respectively. speed and improve the accuracy of image restoration. The processed image of the invention has the advantages of good restoration effect, high definition of the restored image, convenient use, low cost and the like.

Description

technical field [0001] The present invention relates to deep learning and computer vision, in particular to a method for repairing and reproducing old films based on deep learning. Background technique [0002] Film cultural heritage is a precious memory of a country and a nation, an important part of the future intangible cultural heritage, and an excellent carrier for modern Chinese culture to go global. For traditional red movies and movies that reflect the positive spirit of hard work in modern China, modern technology can be used to restore and present them more fully. However, due to the limitations of shooting technology in the past, a large number of old films can no longer meet people's viewing needs for high-definition vision. [0003] There are a huge number of film reels in my country that need to be restored. There are 20,000 to 30,000 feature films in existence, but now only about 60 old film reels can be repaired each year. According to the current restorati...

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): G06T5/00G06T3/40
CPCG06T3/4053G06T2207/10016G06T5/73G06T5/70
Inventor 赵岩聂可卉
Owner 福建帝视科技集团有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com