Edge Computing Application： Real-Time Face Recognition Based on Cloudlet

最新推荐文章于 2024-06-19 13:43:31 发布
最新推荐文章于 2024-06-19 13:43:31 发布
阅读量1.9k 收藏
点赞数 1
分类专栏： 边缘计算 物联网 分布式 边缘计算与物联网 文章标签： 人脸识别 云计算 边缘计算 Cloudlet
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/xingjiarong/article/details/77970021
版权
本文探讨了在移动设备上实现人脸识别的挑战，并提出Cloudlet作为解决方案。Cloudlet通过预处理减小上传数据量，缓存常用数据，动态任务分配以减少响应时间和网络延迟，提升实时人脸识别的用户体验。
摘要由CSDN通过智能技术生成

A mobile-cloud architecture provides a practical platform for performing face recognition on a mobile device. Firstly, even though CPU, memory in mobile phones have been greatly improved, resources demanding applications like face recognition still cannot run on mobile phones. Besides, if send face images to clouds, network latency will not be accepted. However, cloudlet provides an accelerator for face recognition and lessens compute burden on mobile phones which have limited computational capabilities and battery life.

I、Background

人脸检测和识别在日常生活中的应用日渐广泛，例如：安全监控，人脸识别登录，刷脸支付等。随着智能手机的发展，手机APP成为了人脸识别使用最广泛的地方。在过去的十几年间，智能手机经历了一段飞速发展的时期，触屏显示器的使用、从单核处理器到多和处理器、存储从几百M扩展到几百G，这些技术的提高和进步使得智能手机可以运行更加复杂的应用。然而即使如此这样，智能手机的计算和存储能力仍然有限，再加上有限的电池电量，所以一些计算密集型的应用仍然不能在手机上运行，例如：实施人脸检测和识别。其中一种解决方案就是基于云端的方法，将

最低0.47元/天 解锁文章
xingjiarong
关注
专栏目录
【图像拼接】源码精读：Parallax-Tolerant Image Stitching Based on Robust Elastic Warping（ELA/REW）
畅游计算机视觉的海洋
02-05 1万+
论文题目：Parallax-Tolerant Image Stitching Based on Robust Elastic Warping —— 基于非常灵活扭曲的大视差图像拼接算法注：matlab代码。【图像拼接论文精读】专栏对应文章【图像拼接】论文精读：Parallax-Tolerant Image Stitching Based on Robust Elastic Warping（ELA/REW）配合对应的文章阅读，效果更佳！注：请重点关注代码段中的注释！！！
Real-Time-Face-Recognition:使用 OpenCV 进行实时人脸识别
05-29
实时人脸识别 (OpenCV) 这是一个完整的 Python 程序，用于创建高效的实时人脸识别应用程序。 脚步： cmd: python face_taker.py 使用face_taker.py脚本拍照。 输入 ID 号后，脚本将在images文件夹中保存 30 张您的脸部images （必须是整数和增量（从 1 开始，然后是 2、3、...） 注意：确保您的脸部居中。窗口将当所有 30 张照片都拍完后折叠起来。 cmd: python face_train.py face_tain.py脚本将训练一个模型来识别使用face_taker.py脚本拍摄的 30 张图像中的所有人脸，并将训练输出保存在training.yml文件中。 cmd: python face_recognizer.py face_recognizer.py是主要脚本。 您需要在face_taker.py
real-time-face-recognition-with-facenet:只需使用Facenet创建实时人脸识别
05-09
Facenet实时人脸识别 我记得工作的第一天，我被分配去从事人脸识别系统的工作，但是那时做一个能够很好地完成分类器的梦想就像是一个梦，我当时纯粹是使用Open-Cv来检测人脸然后为每张脸创建一个唯一的向量。 但是它的准确性太低，无法像在任何应用程序中一样使用它。 几个月前，我读了一篇名为《 FaceNet：人脸识别和聚类的统一嵌入》的论文，该论文提出了一个用于人脸验证的统一系统。 Facenet是基于使用深度卷积网络学习每个图像的欧几里得嵌入而设计的，嵌入算法可搜索数据的很多维连续表示。 训练网络，使嵌入空间中的平方L2距离直接对应于人脸相似度。 同一个人的脸距离很小，而不同的人的脸距离很大。 一旦产生了嵌入，上述任务就变得直截了当：面部验证仅涉及对两个嵌入之间的距离进行阈值化； 识别成为一个k-NN分类问题，可以使用现成的技术（例如k均值或凝聚聚类）来实现聚类。
Real-Time Face Recognition: An End-to-End Project
jaccen的专栏
08-28 860
Fromhttps://www.hackster.io/mjrobot/real-time-face-recognition-an-end-to-end-project-a10826 We will learn step by step, how to use a PiCam to recognize faces in real-time. IntermediateFull instructions provided12 hours213,376 Things used in this pr...
【啃论文俱乐部】雾计算的编排
贺贺her的博客
04-13 393
啃论文俱乐部--雾计算的编排
如何理解Cloudlet
aifeng2014的专栏
02-01 2178
cloudlet 的理解
Practical Real-time Data Processing and Analytics epub
10-11
Practical Real-time Data Processing and Analytics 英文epub 本资源转载自网络，如有侵权，请联系上传者或csdn删除 本资源转载自网络，如有侵权，请联系上传者或csdn删除
Real-Time-Facial-Expression-Recognition-with-DeepLearning
12-27
Real-Time-Facial-Expression-Recognition-with-DeepLearning
人脸识别face-recognition
03-14
This repository contains the code for the face recognition system developed by the FACE creative inquiry. We are developing an accelerated, real-time recognition system based on several popular face recognition techniques.含有三个文件夹和说明
Cloud and Fog Computing in 5G Mobile Networks-IET(2017).pdf
02-09
In order to continue to ensure the sustainability of mobile communication services over the next decade and to meet the business and consumer demands, fifth genera- tion (5G) mobile communication services is expected to be rolled out by 2020. One of the major requirements for 5G networks is the significant spectral efficiency (SE) enhancement compared to fourth generation (4G) as the anticipated exponential increase in the volume of mobile data traffic is huge, for example, at least 1,000-fold in the 2020s compared to 2010. In particular, the peak data rate in 5G should be 10–20 Gbps that is 10–20 times the peak data rate in 4G, and the user experienced data rate should be 1 Gbps (100 times the user experienced data rate in 4G). In addition, the rapid development of Mobile Internet and the Internet of Things (IoT) exponentially accelerates the demands for high data rate applications, including high-quality video streaming, social networking, and machine-to-machine communications. In cellular network, the design of radio access technology, in general, and multiple access technique, in particular, are one of the most important aspects in improving the system capacity. Multiple access techniques are usually categorized into two orthogonal and nonorthogonal approaches [1]. In orthogonal approaches, 2 Cloud and fog computing in 5G mobile networks signals from different users are orthogonal to each other, that is, their cross correlation is zero (the available resources such as the system bandwidth (BW), and time is divided among users). Nonorthogonal schemes such as code division multiple access (CDMA) allow nonzero cross correlation among the signals from different users. Second and third generation cellular systems such as IS-95, CDMA2000, and wideband-CDMA (WCDMA) have adopted nonorthogonal multiple access (NOMA) techniques. CDMA is usually more robust against fading and cross-cell interference, but is susceptible to intracell interference. With careful cell planning, orthogonal multiple access (OMA) can avoid intracell interference. On that, most of the first and second generation cellular systems adopted orthogonal MA approaches. Even, orthogonal frequency division multiple access–based OMA has been adopted in 4G systems such as long-term evolution (LTE) and LTE-advanced. Despite a practical advantage of intracell interference avoidance capability, CDMA has limited data rate due to its spread-spectrum nature. OMA is a realistic choice for achieving good performance in terms of system-level throughput. However, due to the aforementioned upcoming wave, 5G networks require further enhancement in the system efficacy. Then again, to get the facilities of on-demand resource processing, delay-aware storage, and high network capacity, the cloud computing–based radio access infrastructure is a possible solution. And, advanced baseband computation and radio frequency communication are required to enable large-scale cooperative signal processing in the physical layer and adapted to new air interfaces in 5G systems. In this regard, researchers over the globe have started investigating NOMA as a promising multiple access scheme for future radio access. NOMA achieves superior spectral efficiencies by combining superposition coding (SC) at the transmitter with successive interference cancelation (SIC) at the recei- vers [2,3]. On the top of that, the evolution of wireless networks into 5G poses new challenges on energy efficiency (EE), as the entire network will be ultradense. With an extreme increase in number of infrastructure nodes, the total energy consump- tion may simply surpass an acceptable level. Although the substantial energy is basically consumed by the hardware, the NOMA has an inherent ability to adapt the transmission strategy according to the traffic and users channel state information (CSI). Thus, it can achieve a good operating point where both the spectrum efficiency and EE become optimum. In view of the fact that the IoT is expected to be widely used in our everyday life, the fog computing is growing in popularity. One of the primary objectives of fog networking is minimizing the use of BW. Although fog computing is implemented by handing some application services at edge devices and in a remote data center, some physical and medium access control layer issues can help to achieve its efficient spectrum utilization intention. In this regard, NOMA is important, as its target is also the efficient utilization of available spectrum. Over the past few years, NOMA has attracted huge attention of researchers to meet the 5G requirements. As a consequence, many research efforts on this field already exist. Research trends in NOMA include diverse topics, for example, performance analysis, cooperative communications, and fairness analysis. However, NOMA in 5G is still in its infancy. At this stage, a comprehensive knowledge on the up-to-date research status of NOMA in 5G systems is extremely useful to researchers to do more research in this area. In this chapter, we appraise the state of the art of NOMA schemes for 5G green mobile networks 3 NOMA research trends and disclose various issues that need to be addressed to transform radio access techniques through NOMA innovation.
关于chrome、edge浏览器f12开发者模式的application中无法添加参数的问题
Doubletree.lin的博客
02-09 5975
chrome、edge的开发这模式无法添加applictioin参数
微软Edge浏览器开发者工具的全面探索
最新发布
2401_85439108的博客
06-19 989
微软Edge浏览器的开发者工具为Web开发提供了全方位的支持，从基础的元素检查到高级的性能分析，再到移动设备模拟和安全性检查，这些工具极大地提高了开发效率和网页质量。随着Web技术的不断发展，Edge开发者工具也在不断更新和完善，为开发者提供更多的功能和更好的体验。微软Edge浏览器的开发者工具是一套功能强大的网页调试和分析工具，为前端开发者和网站管理员提供了深入理解、测试和优化网页的能力。从简单的元素检查到复杂的性能分析，Edge的开发者工具涵盖了Web开发的多个方面。
实时人脸检测 (Real-Time Face Detection)
weixin_30480651的博客
03-22 469
源地址：http://blog.sina.com.cn/s/blog_79b67dfe0102uzra.html 最近需要用到人脸检测，于是找了篇引用广泛的论文实现了一下：Robust Real-Time Face Detection。实现的过程主要有三个步骤：人脸数据准备，算法实现，算法调试。 人脸数据集的准备：网上有很多免费的和付费的。比如这里有个网页介绍了一些常用的人脸数据库。我...
计算机网络实验_热烈祝贺！实验室四项成果被计算机网络领域顶级学术会议INFOCOM 2020 录用...
weixin_39598584的博客
12-15 698
近日，计算机网络领域顶级学术会议INFOCOM 2020(IEEE International Conference on Computer Communications)论文录用名单出炉，天津市先进网络技术与应用重点实验室4篇论文被该会议录用。 IEEE INFOCOM在计算机网络科研领域与MOBICOM和SIGCOMM并称三大顶级学术会议，是世界顶尖科研...
边缘计算Edge Computing
weixin_42631919的博客
08-07 2674
最近工作有接触到这个概念，先贴点资料在这里吧。 Overview Edge computing pushes applications, data and computing power (services) away from centralized points to the logical extremes of a network. Edge computing takes advan...
卸载Win10新Edge浏览器
dyk2200364978dyk的博客
12-17 655
进入目录 C:\Program Files (x86)\Microsoft\Edge\Application\87.0.664.60\Installer 单击选中 setup.exe 后 进入powershell 执行命令 .\setup.exe --uninstall --system-level --verbose-logging --force-uninstall 即可完成卸载
Java for High-Performance Computing: MPI-Based Solutions and Integration
"Java for High Performance Computing (HPC) 是一门专门针对高性能计算领域而设计的Java技术课程。该课程主要探讨如何利用Java语言与高性能计算平台上的Message Passing Interface (MPI)标准进行集成，以提升在...
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值