当前位置:网站首页>Reference frame generation based on deep learning
Reference frame generation based on deep learning
2022-07-06 20:56:00 【Dillon2015】
This article comes from the proposal JVET-T0058 and JVET-U0087, This method generates virtual reference frames for inter frame prediction by inserting frames . The whole model consists of several sub models , Perform optical flow estimation respectively 、 Compensation and detail enhancement .
The overall architecture
The overall architecture is as follows Fig.1 Shown , In the process of video coding DPB There is a reference frame for motion estimation , according to GOP Structure the current frame has one or more forward 、 Backward reference frame . The default in the proposal is POC The two reference frames closest to the current frame generate a virtual reference frame , Such as Fig.1 Current frame in POC yes 5, Then use POC by 4 and 6 The frame of generates a reference frame . The generated virtual reference frame will be put into DPB For reference , Virtual reference frame POC Set to the same as the current frame . In order to prevent affecting the time domain MVP According to the POC Distant MV Zoom process , Virtual reference frame MV All set to 0 And is used as a long-term reference frame . In the proposal , After the current frame is decoded, the virtual reference frame starts from DPB Remove .
For high resolution sequences (4K or 8K) Due to resource constraints, neural network processing cannot be directly used for the whole frame , At this time, it is assumed that the virtual reference frame is divided into multiple regions , Each area uses network generation separately , Then put these areas together into a reference frame .
A network model
Optical flow estimation and compensation are mostly used in general video interpolation , Generally, bidirectional optical flow method is used , Then the two optical flows are combined into one through a linear model . Only the single optical flow model is used in the proposal .
Such as Fig.2, First, optical flow is generated by optical flow estimation model ( Input is POC The two nearest reference frames ), And then through backward warping Process processing optical flow , The processed optical flow and two reference frames pass through fusion Process synthesis intermediate frame . The intermediate frame will enhance the quality of the model through details , The detail enhancement model consists of two parts ,PCD(Pyramid, Cascading and Deformable) For space-time optimization and TSA (Temporal and Spatial Attention) Used to improve important features attention.
experimental result
Interested parties, please pay attention to WeChat official account Video Coding
边栏推荐
- Tips for web development: skillfully use ThreadLocal to avoid layer by layer value transmission
- C language operators
- Build your own application based on Google's open source tensorflow object detection API video object recognition system (IV)
- No Yum source to install SPuG monitoring
- 7. Data permission annotation
- SSO single sign on
- [wechat applet] operation mechanism and update mechanism
- Web开发小妙招:巧用ThreadLocal规避层层传值
- Intel 48 core new Xeon run point exposure: unexpected results against AMD zen3 in 3D cache
- 1500萬員工輕松管理,雲原生數據庫GaussDB讓HR辦公更高效
猜你喜欢
SAP UI5 框架的 manifest.json
Distributed ID
The most comprehensive new database in the whole network, multidimensional table platform inventory note, flowus, airtable, seatable, Vig table Vika, flying Book Multidimensional table, heipayun, Zhix
The mail command is used in combination with the pipeline command statement
【mysql】触发器
[DIY]如何制作一款个性的收音机
2022 Guangdong Provincial Safety Officer C certificate third batch (full-time safety production management personnel) simulation examination and Guangdong Provincial Safety Officer C certificate third
知识图谱之实体对齐二
Kubernetes learning summary (20) -- what is the relationship between kubernetes and microservices and containers?
防火墙基础之外网服务器区部署和双机热备
随机推荐
SAP Fiori应用索引大全工具和 SAP Fiori Tools 的使用介绍
APS taps home appliance industry into new growth points
What programming do children learn?
Common doubts about the introduction of APS by enterprises
User defined current limiting annotation
Pat 1085 perfect sequence (25 points) perfect sequence
New database, multidimensional table platform inventory note, flowus, airtable, seatable, Vig table Vika, Feishu multidimensional table, heipayun, Zhixin information, YuQue
Quel genre de programmation les enfants apprennent - ils?
OAI 5g nr+usrp b210 installation and construction
Design your security architecture OKR
【微信小程序】運行機制和更新機制
How to turn a multi digit number into a digital list
【每周一坑】计算100以内质数之和 +【解答】输出三角形
Pycharm remote execution
快过年了,心也懒了
How to upgrade high value-added links in the textile and clothing industry? APS to help
Implementation of packaging video into MP4 format and storing it in TF Card
OAI 5G NR+USRP B210安装搭建
Minimum cut edge set of undirected graph
华为设备命令