基于深度学习的ARM平台实时人脸识别

doi:10.11772/j.issn.1001-9081.2019010164

计算机应用 ›› 2019, Vol. 39 ›› Issue (8): 2217-2222.DOI: 10.11772/j.issn.1001-9081.2019010164

基于深度学习的ARM平台实时人脸识别

方国康^1,2, 李俊^1,2, 王垚儒^1,2

1. 武汉科技大学计算机科学与技术学院, 武汉 430065;
2. 智能信息处理与实时工业系统湖北省重点实验室(武汉科技大学), 武汉 430065

收稿日期:2019-01-22 修回日期:2019-03-14 出版日期:2019-08-10 发布日期:2019-04-15
通讯作者: 李俊
作者简介:方国康(1994-),男,湖北恩施人,硕士研究生,主要研究方向:智能计算、计算机视觉;李俊(1978-),男,湖北黄石人,副教授,博士,主要研究方向:智能计算、机器学习;王垚儒(1995-);男,湖北天门人,硕士研究生,主要研究方向:智能计算、深度学习。
基金资助:
国家自然科学基金资助项目（61572381）；武汉科技大学智能信息处理与实时工业系统湖北重点实验室基金资助项目（znxx2018QN06）。

Real-time face recognition on ARM platform based on deep learning

FANG Guokang^1,2, LI Jun^1,2, WANG Yaoru^1,2

1. School of Computer Science and Technology, Wuhan University of Science and Technology, Wuhan Hubei 430065, China;
2. Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System(Wuhan University of Science and Technology), Wuhan Hubei 430065, China

Received:2019-01-22 Revised:2019-03-14 Online:2019-08-10 Published:2019-04-15
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61572381), the Hubei Key Laboratory of Intelligent Information Processing and Real-time Industrial System Foundation of Wuhan University of Science and Technology (znxx2018QN06).

摘要/Abstract

摘要： 针对ARM平台上人脸识别实时性不强和识别率低的问题，提出一种基于深度学习的实时人脸识别方法。首先基于MTCNN人脸检测算法设计了一种实时检测并追踪人脸的算法；然后在ARM平台上基于深度残差网络（ResNet）设计人脸特征提取网络；最后针对ARM平台的特点，使用Mali-GPU加速人脸特征提取网络的运算，分担CPU负荷，提高系统整体运行效率。算法部署在基于ARM的瑞芯微RK3399开发板上，运行速度达到22 帧/s。实验结果表明，与MobileFaceNet相比，该方法在MegaFace上的识别率提升了11个百分点。

关键词: ARM平台, 人脸识别, 人脸追踪, 残差网络, Mali-GPU

Abstract: Aiming at the problem of low real-time performance of face recognition and low face recognition rate on ARM platform, a real-time face recognition method based on deep learning was proposed. Firstly, an algorithm for detecting and tracking faces in real time was designed based on MTCNN face detection algorithm. Then, a face feature extraction network was designed based on Residual Neural Network (ResNet) on ARM platform. Finally, according to the characteristics of ARM platform, Mali-GPU was used to accelerate the operation of face feature extraction network, sharing the CPU load and improving the overall running efficiency of the system. The algorithm was deployed on ARM-based Rockchip development board, and the running speed reaches 22 frames per second. Experimental results show that the recognition rate of this method is 11 percentage points higher than that of MobileFaceNet on MegaFace.

Key words: ARM platform, face recognition, face tracking, Residual neural Network (ResNet), Mali-GPU

中图分类号:

方国康, 李俊, 王垚儒. 基于深度学习的ARM平台实时人脸识别[J]. 计算机应用, 2019, 39(8): 2217-2222.

FANG Guokang, LI Jun, WANG Yaoru. Real-time face recognition on ARM platform based on deep learning[J]. Journal of Computer Applications, 2019, 39(8): 2217-2222.

参考文献

[1] 刘飞,张俊然,杨豪.基于深度学习的糖尿病患者的分类识别[J].计算机应用,2018,38(S1):39-43. (LIU F, ZHANG J R, YANG H. Classification and recognition of diabetes mellitus based on deep learning[J]. Journal of Computer Applications, 2018, 38(S1):39-43.)
[2] 熊咏平,丁胜,邓春华,等.基于深度学习的复杂气象条件下海上船只检测[J].计算机应用,2018,38(12):3631-3637. (XIONG Y P, DING S, DENG C H, et al. Ship detection under complex sea and weather conditions based on deep learning[J]. Journal of Computer Applications, 2018, 38(12):3631-3637.)
[3] 孙劲光,孟凡宇.基于深度神经网络的特征加权融合人脸识别方法[J].计算机应用,2016,36(2):437-443. (SUN J G, MENG F Y. Face recognition based on deep neural network and weighted fusion of face features[J]. Journal of Computer Applications, 2016, 36(2):437-443.)
[4] KAZEMI V, SULLIVAN J. One millisecond face alignment with an ensemble of regression trees[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2014:1867-1874.
[5] DASHORE G, Dr. RAJ V C. An efficient method for face recognition using Principal Component Analysis (PCA)[J]. International Journal of Advanced Technology and Engineering Research, 2012, 2(2):23-29.
[6] CHAN L, SALLEH S, TING C. Face biometrics based on principal component analysis and linear discriminant analysis[J]. Journal of Computer Science, 2010, 6(7):693-699.
[7] PENEV P S, ATICK J J. Local feature analysis:a general statistical theory for object representation[J]. Network Computation in Neural Systems, 1996, 7(3):477-500.
[8] 杨天明,陈志,岳文静.基于视频深度学习的时空双流人物动作识别模型[J]. 计算机应用,2018,38(3):895-899. (YANG T M, CHEN Z, YUE W J. Spatio-temporal two-stream human action recognition model based on video deep learning[J]. Journal of Computer Applications, 2018, 38(3):895-899.)
[9] LECUN Y L, BOTTOU L, BENGIO Y,et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11):2278-2324.
[10] TAIGMAN Y, YANG M, RANZATO M, et al. DeepFace:closing the gap to human-level performance in face verification[C]//Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2014:1701-1708.
[11] HUANG G B, RAMESH M, BERG T, et al. Labeled faces in the wild:a database for studying face recognition in unconstrained environments[EB/OL].[2018-12-05]. http://cs.umass.edu/~elm/papers/lfw.pdf.
[12] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2016:770-778.
[13] SIMONVAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition[J]. arXiv E-print, 2015:arXiv:1409.1556.
[14] SZEGEDY C, LIU W, JIA Y, et al. Going deeper with convolutions[C]//Proceedings of the 2015 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2015:1-9.
[15] ZHANG K, ZHANG Z, LI Z, et al. Joint face detection and alignment using multitask cascaded convolutional networks[J]. IEEE Signal Processing Letters, 2016, 23(10):1499-1503.
[16] LIU W, WEN Y, YU Z, et al. SphereFace:deep hypersphere embedding for face recognition[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2017:6738-6746.
[17] KEMELMACHER-SHLIZERMAN I, SEITZ S M, MILLER D,et al. The MegaFace benchmark:1 million faces for recognition at scale[C]//Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Washington, DC:IEEE Computer Society, 2016:4873-4882.
[18] WANG F, XIANG X, CHENG J, et al. NormFace:L2 hypersphere embedding for face verification[J]. arXiv E-print, 2017:arXiv:1704.06369.
[19] WANG F, LIU W, LIU H, et al. Additive margin softmax for face verification[J]. arXiv E-print, 2018:arXiv:1801.05599.
[20] DENG J, GUO J, XUE N, et al. ArcFace:additive angular margin loss for deep face recognition[J]. arXiv E-print, 2019:arXiv:1801.07698.
[21] GRASSO I, RADOJKOVIC P, RAJOVIC N, et al. Energy efficient HPC on embedded SoCs:Optimization techniques for mali GPU[C]//Proceedings of the 2014 IEEE International Parallel and Distributed Processing Symposium. Washington, DC:IEEE Computer Society, 2014:123-132.
[22] HOWARD A G, ZHU M, CHEN B, et al. MobileNets:efficient convolutional neural networks for mobile vision applications[J]. arXiv E-print, 2017:arXiv:1704.04861.
[23] ZHANG X, ZHOU X, LIN M, et al. ShuffleNet:an extremely efficient convolutional neural network for mobile devices[J]. arXiv E-print, 2017:arXiv:1707.01083v1.
[24] CHEN T, MOREAU T, JIANG Z, et al. TVM:an automated end-to-end optimizing compiler for deep learning[J]. arXiv E-print, 2018:arXiv:1802.04799.
[25] CHEN S, LIU Y, GAO X, et al. MobileFaceNets:efficient CNNs for accurate real-time face verification on mobile devices[J]. arXiv E-print, 2018:arXiv:1804.07573.
[26] CHOLLET F. Xception:deep learning with depthwise separable convolutions[J]. arXiv E-print, 2016:arXiv:1610.02357v2.
[27] SANDLER M, HOWARD A, ZHU M, et al. MobileNetV2:inverted residuals and linear bottlenecks[J]. arXiv E-print, 2019:arXiv:1801.04381.
[28] GUO Y, ZHANG L, HU Y, et al. MS-Celeb-1M:a dataset and benchmark for large-scale face recognition[J]. arXiv E-print, 2016:arXiv:1607.08221.
[29] SENGUPTA S, CHEN J, CASTILLO C, et al. Frontal to profile face verification in the wild[C]//Proceedings of the 2016 IEEE Winter Conference on Applications of Computer Vision. Washington, DC:IEEE Computer Society, 2016:1-9.
[30] MOSCHOGLOU S, PAPAIOANNOU A, SAGONAS C, et al. AgeDB:the first manually collected, in-the-wild age database[C]//Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops. Washington, DC:IEEE Computer Society, 2017:1997-2005.

基于深度学习的ARM平台实时人脸识别

Real-time face recognition on ARM platform based on deep learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	王贺兵, 张春梅. 基于非对称卷积-压缩激发-次代残差网络的人脸关键点检测[J]. 计算机应用, 2021, 41(9): 2741-2747.
[2]	高工, 杨红雨, 刘洪. 基于深度学习的三维点云人脸识别[J]. 计算机应用, 2021, 41(9): 2736-2740.
[3]	刘世泽, 朱奕达, 陈润泽, 罗海勇, 赵方, 孙艺, 王宝会. 基于残差时域注意力神经网络的交通模式识别算法[J]. 计算机应用, 2021, 41(6): 1557-1565.
[4]	刘世泽, 秦艳君, 王晨星, 高存远, 罗海勇, 赵方, 王宝会. 基于多尺度特征提取的交通模式识别算法[J]. 计算机应用, 2021, 41(6): 1573-1580.
[5]	王永金, 左羽, 吴恋, 崔忠伟, 赵晨洁. 基于注意力机制的图像超分辨率重建[J]. 计算机应用, 2021, 41(3): 845-850.
[6]	任奕茗, 王让定, 严迪群, 林昱臻. 基于深度残差网络的语音隐写分析方法[J]. 计算机应用, 2021, 41(3): 774-779.
[7]	李虹霞, 秦品乐, 闫寒梅, 曾建潮, 鲍骞月, 柴锐. 基于面部特征图对称的人脸正面化生成对抗网络算法[J]. 计算机应用, 2021, 41(3): 714-720.
[8]	佘玉龙, 张晓龙, 程若勤, 邓春华. 基于边缘关注模型的语义分割方法[J]. 计算机应用, 2021, 41(2): 343-349.
[9]	钟莎, 黄玉清. 基于孪生区域候选网络的无人机指定目标跟踪[J]. 计算机应用, 2021, 41(2): 523-529.
[10]	陈朗, 王让定, 严迪群, 林昱臻. 融合残差网络和极限梯度提升的音频隐写检测模型[J]. 计算机应用, 2021, 41(2): 449-455.
[11]	方书雅, 刘守印. 基于学生人体检测的无感知课堂考勤方法[J]. 计算机应用, 2020, 40(9): 2519-2524.
[12]	赵青, 余元辉. 基于分层特征化网络的三维人脸识别[J]. 计算机应用, 2020, 40(9): 2514-2518.
[13]	闵鑫, 王海鹏, 牟长宁. 基于多头注意力机制和残差神经网络的肽谱匹配打分算法[J]. 计算机应用, 2020, 40(6): 1830-1836.
[14]	代强, 程曦, 王永梅, 牛子未, 刘飞. 基于轻量自动残差缩放网络的图像超分辨率重建[J]. 计算机应用, 2020, 40(5): 1446-1452.
[15]	边小勇, 费雄君, 穆楠. 基于尺度注意力网络的遥感图像场景分类[J]. 计算机应用, 2020, 40(3): 872-877.