Video processing method and apparatus

By using lightweight residual dense blocks and cascaded convolutional neural networks in video processing, the problem of low parameter transmission efficiency in video super-resolution is solved, and efficient video restoration is achieved.

CN114096987BActive Publication Date: 2025-11-11KOREA ADVANCED INST OF SCI & TECH
2 Cites 0 Cited by

Patent Information

Application Number
CN202080043521.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2020-07-01
Filing Date
2020-07-03
Publication Date
2025-11-11
Estimated Expiration
2040-07-03

AI Technical Summary

Technical Problem

Existing convolutional neural networks suffer from low parameter transmission efficiency in video super-resolution processing, especially at receivers with limited bandwidth, where they struggle to effectively recover high-quality video.

Method used

By employing lightweight residual dense blocks and cascaded convolutional neural networks, video super-resolution, inverse tone mapping, frame interpolation, motion deblurring, denoising, and compression artifact removal are achieved by segmenting video frames and learning residual parameters, thereby reducing the amount of data transmitted and improving the quality of video restoration.

Benefits of technology

It effectively reduces the amount of data transmitted while improving the quality and efficiency of video recovery, making it suitable for receiving devices with limited bandwidth.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

This invention discloses a video processing method and apparatus. According to one embodiment, the video processing method includes the following steps: receiving a video comprising multiple temporal portions; receiving first model parameters corresponding to a first neural network used for overall processing of the video; receiving residuals between the first model parameters and multiple second model parameters corresponding to multiple second neural networks used to process the multiple temporal portions respectively; and performing at least one of super-resolution, inverse tone mapping, tone mapping, frame interpolation, motion deblurring, denoising, and compression artifact removal on the video based on the residuals.
Need to check novelty before this filing date? Find Prior Art