Tamed Warping Network for High-Resolution Semantic Video Segmentation

IF 2.5 4区 综合性期刊 Q2 CHEMISTRY, MULTIDISCIPLINARY Applied Sciences-Basel Pub Date : 2023-09-07 DOI:10.3390/app131810102
Songyuan Li, Junyi Feng, Xi Li
{"title":"Tamed Warping Network for High-Resolution Semantic Video Segmentation","authors":"Songyuan Li, Junyi Feng, Xi Li","doi":"10.3390/app131810102","DOIUrl":null,"url":null,"abstract":"Recent approaches for fast semantic video segmentation have reduced redundancy by warping feature maps across adjacent frames, greatly speeding up the inference phase. However, the accuracy drops seriously owing to the errors incurred by warping. In this paper, we propose a novel framework and design a simple and effective correction stage after warping. Specifically, we build a non-key-frame CNN, fusing warped context features with current spatial details. Based on the feature fusion, our context feature rectification (CFR) module learns the model’s difference from a per-frame model to correct the warped features. Furthermore, our residual-guided attention (RGA) module utilizes the residual maps in the compressed domain to help CRF focus on error-prone regions. Results on Cityscapes show that the accuracy significantly increases from 67.3% to 71.6%, and the speed edges down from 65.5 FPS to 61.8 FPS at a resolution of 1024×2048. For non-rigid categories, e.g., “human” and “object”, the improvements are even higher than 18 percentage points.","PeriodicalId":48760,"journal":{"name":"Applied Sciences-Basel","volume":null,"pages":null},"PeriodicalIF":2.5000,"publicationDate":"2023-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Sciences-Basel","FirstCategoryId":"103","ListUrlMain":"https://doi.org/10.3390/app131810102","RegionNum":4,"RegionCategory":"综合性期刊","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"CHEMISTRY, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0

Abstract

Recent approaches for fast semantic video segmentation have reduced redundancy by warping feature maps across adjacent frames, greatly speeding up the inference phase. However, the accuracy drops seriously owing to the errors incurred by warping. In this paper, we propose a novel framework and design a simple and effective correction stage after warping. Specifically, we build a non-key-frame CNN, fusing warped context features with current spatial details. Based on the feature fusion, our context feature rectification (CFR) module learns the model’s difference from a per-frame model to correct the warped features. Furthermore, our residual-guided attention (RGA) module utilizes the residual maps in the compressed domain to help CRF focus on error-prone regions. Results on Cityscapes show that the accuracy significantly increases from 67.3% to 71.6%, and the speed edges down from 65.5 FPS to 61.8 FPS at a resolution of 1024×2048. For non-rigid categories, e.g., “human” and “object”, the improvements are even higher than 18 percentage points.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
用于高分辨率语义视频分割的Tamed Warping网络
最近的快速语义视频分割方法通过在相邻帧之间扭曲特征图来减少冗余,大大加快了推理阶段。然而,由于翘曲引起的误差,精度严重下降。在本文中,我们提出了一种新的框架,并设计了一个简单有效的翘曲后矫正台。具体来说,我们构建了一个非关键帧CNN,将扭曲的上下文特征与当前的空间细节融合在一起。基于特征融合,我们的上下文特征校正(CFR)模块学习模型与每帧模型的差异,以校正扭曲的特征。此外,我们的残差引导注意力(RGA)模块利用压缩域中的残差图来帮助CRF关注容易出错的区域。在Cityscapes上的结果显示,在1024×2048的分辨率下,准确率从67.3%显著提高到71.6%,速度从65.5 FPS下降到61.8 FPS。对于非刚性类别,例如“人”和“物体”,改进甚至超过18个百分点。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Applied Sciences-Basel
Applied Sciences-Basel CHEMISTRY, MULTIDISCIPLINARYMATERIALS SCIE-MATERIALS SCIENCE, MULTIDISCIPLINARY
CiteScore
5.30
自引率
11.10%
发文量
10882
期刊介绍: Applied Sciences (ISSN 2076-3417) provides an advanced forum on all aspects of applied natural sciences. It publishes reviews, research papers and communications. Our aim is to encourage scientists to publish their experimental and theoretical results in as much detail as possible. There is no restriction on the length of the papers. The full experimental details must be provided so that the results can be reproduced. Electronic files and software regarding the full details of the calculation or experimental procedure, if unable to be published in a normal way, can be deposited as supplementary electronic material.
期刊最新文献
Application of Digital Holographic Imaging to Monitor Real-Time Cardiomyocyte Hypertrophy Dynamics in Response to Norepinephrine Stimulation. Reproducibility in Radiomics: A Comparison of Feature Extraction Methods and Two Independent Datasets. Study on Shear Resistance and Structural Performance of Corrugated Steel–Concrete Composite Deck Clustering Analysis of Wind Turbine Alarm Sequences Based on Domain Knowledge-Fused Word2vec Unraveling Functional Dysphagia: A Game-Changing Automated Machine-Learning Diagnostic Approach
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1