Historical Blurry Video-Based Face Recognition.

IF 2.7 Q3 IMAGING SCIENCE & PHOTOGRAPHIC TECHNOLOGY Journal of Imaging Pub Date : 2024-09-20 DOI:10.3390/jimaging10090236
Lujun Zhai, Suxia Cui, Yonghui Wang, Song Wang, Jun Zhou, Greg Wilsbacher
{"title":"Historical Blurry Video-Based Face Recognition.","authors":"Lujun Zhai, Suxia Cui, Yonghui Wang, Song Wang, Jun Zhou, Greg Wilsbacher","doi":"10.3390/jimaging10090236","DOIUrl":null,"url":null,"abstract":"<p><p>Face recognition is a widely used computer vision, which plays an increasingly important role in user authentication systems, security systems, and consumer electronics. The models for most current applications are based on high-definition digital cameras. In this paper, we focus on digital images derived from historical motion picture films. Historical motion picture films often have poorer resolution than modern digital imagery, making face detection a more challenging task. To approach this problem, we first propose a trunk-branch concatenated multi-task cascaded convolutional neural network (TB-MTCNN), which efficiently extracts facial features from blurry historical films by combining the trunk with branch networks and employing various sizes of kernels to enrich the multi-scale receptive field. Next, we build a deep neural network-integrated object-tracking algorithm to compensate for failed recognition over one or more video frames. The framework combines simple online and real-time tracking with deep data association (Deep SORT), and TB-MTCNN with the residual neural network (ResNet) model. Finally, a state-of-the-art image restoration method is employed to reduce the effect of noise and blurriness. The experimental results show that our proposed joint face recognition and tracking network can significantly reduce missed recognition in historical motion picture film frames.</p>","PeriodicalId":37035,"journal":{"name":"Journal of Imaging","volume":"10 9","pages":""},"PeriodicalIF":2.7000,"publicationDate":"2024-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11433217/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Imaging","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3390/jimaging10090236","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"IMAGING SCIENCE & PHOTOGRAPHIC TECHNOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

Face recognition is a widely used computer vision, which plays an increasingly important role in user authentication systems, security systems, and consumer electronics. The models for most current applications are based on high-definition digital cameras. In this paper, we focus on digital images derived from historical motion picture films. Historical motion picture films often have poorer resolution than modern digital imagery, making face detection a more challenging task. To approach this problem, we first propose a trunk-branch concatenated multi-task cascaded convolutional neural network (TB-MTCNN), which efficiently extracts facial features from blurry historical films by combining the trunk with branch networks and employing various sizes of kernels to enrich the multi-scale receptive field. Next, we build a deep neural network-integrated object-tracking algorithm to compensate for failed recognition over one or more video frames. The framework combines simple online and real-time tracking with deep data association (Deep SORT), and TB-MTCNN with the residual neural network (ResNet) model. Finally, a state-of-the-art image restoration method is employed to reduce the effect of noise and blurriness. The experimental results show that our proposed joint face recognition and tracking network can significantly reduce missed recognition in historical motion picture film frames.

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
基于历史模糊视频的人脸识别。
人脸识别是一种广泛应用的计算机视觉技术,在用户身份验证系统、安全系统和消费电子产品中发挥着越来越重要的作用。目前大多数应用的模型都基于高清数码相机。在本文中,我们将重点关注从历史电影胶片中提取的数字图像。与现代数字图像相比,历史电影胶片的分辨率通常较低,这使得人脸检测成为一项更具挑战性的任务。为了解决这个问题,我们首先提出了一种主干-分支串联多任务级联卷积神经网络(TB-MTCNN),它通过将主干网络与分支网络相结合,并采用不同大小的核来丰富多尺度感受野,从而有效地从模糊的历史影片中提取面部特征。接下来,我们构建了一种深度神经网络集成的对象跟踪算法,以补偿一个或多个视频帧的识别失败。该框架将简单的在线实时跟踪与深度数据关联(Deep SORT)相结合,并将 TB-MTCNN 与残差神经网络(ResNet)模型相结合。最后,还采用了最先进的图像修复方法来减少噪声和模糊的影响。实验结果表明,我们提出的联合人脸识别和跟踪网络可以显著减少历史电影胶片中的漏识现象。
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
Journal of Imaging
Journal of Imaging Medicine-Radiology, Nuclear Medicine and Imaging
CiteScore
5.90
自引率
6.20%
发文量
303
审稿时长
7 weeks
期刊最新文献
A Specialized Pipeline for Efficient and Reliable 3D Semantic Model Reconstruction of Buildings from Indoor Point Clouds. Design and Use of a Custom Phantom for Regular Tests of Radiography Apparatus: A Feasibility Study. Differentiation of Benign and Malignant Neck Neoplastic Lesions Using Diffusion-Weighted Magnetic Resonance Imaging. Investigating the Sim-to-Real Generalizability of Deep Learning Object Detection Models. Quantitative Comparison of Color-Coded Parametric Imaging Technologies Based on Digital Subtraction and Digital Variance Angiography: A Retrospective Observational Study.
×
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