Video decoding method and device
By adjusting the number of modules in the loop filter network, a target loop filter network is generated, which solves the problem of insufficient versatility and flexibility of fixed-structure deep neural network models in video coding, and realizes efficient loop filtering that can adapt to different application scenarios.
CN122093564APending Publication Date: 2026-05-26HISENSE VISUAL TECH CO LTD
0 Cites 0 Cited by
Patent Information
- Application Number
- CN202411709172.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-26
- Publication Date
- 2026-05-26
AI Technical Summary
Technical Problem
In existing technologies, fixed-structure deep neural network models lack versatility and flexibility in video coding loop filtering tasks, and cannot adapt to the complexity requirements of loop filters in different application scenarios.
Method used
By adjusting the number of modules in the initial loop filter network according to the obtained filtering strategy, a target loop filter network is generated to adapt to different degradation levels of video frames and application scenario requirements, thereby achieving flexible loop filtering.
Benefits of technology
It achieves greater versatility and flexibility, adapts to the complexity requirements of loop filters in different application scenarios, and improves the quality and efficiency of video frames.
✦ Generated by Eureka AI based on patent content.
Abstract
The invention provides a video decoding method and device, and relates to the technical field of video coding and decoding. Comprising the steps of obtaining coded data of a target video frame; decoding the coded data of the target video frame to obtain a reconstructed video frame of the target video frame; a filtering strategy is obtained, the filtering strategy is that elements are binary vectors of a target number, and the target number is the number of filtering modules of the initial loop filtering network; adjusting the number of filtering modules of the initial loop filtering network based on the filtering strategy to obtain a target loop filtering network; and performing loop filtering on the reconstructed video frame through the target loop filtering network to obtain an enhanced video frame corresponding to the reconstructed video frame. Some embodiments of the application are used for realizing loop filtering with higher universality and flexibility, thereby adapting to the requirements of different application scenarios on the complexity of the loop filter.
Need to check novelty before this filing date? Find Prior Art