|本期目录/Table of Contents|

[1]包晓安,胡玲玲,张娜,等.基于级联网络的快速人脸检测算法[J].浙江理工大学学报,2019,41-42(自科三):347-353.
 BAO Xiaoan,HU Lingling,ZHANG Na,et al.Fast face detection algorithm based on cascade network[J].Journal of Zhejiang Sci-Tech University,2019,41-42(自科三):347-353.
点击复制

基于级联网络的快速人脸检测算法()
分享到:

浙江理工大学学报[ISSN:1673-3851/CN:33-1338/TS]

卷:
第41-42卷
期数:
2019年自科三期
页码:
347-353
栏目:
出版日期:
2019-06-23

文章信息/Info

Title:
Fast face detection algorithm based on cascade network
文章编号:
1673-3851 (2019) 05-0347-07
作者:
包晓安胡玲玲张娜吴彪桂江生
1.浙江理工大学信息学院,杭州 310018; 2.山口大学东亚研究科,日本山口 7538514
Author(s):
BAO Xiaoan HU Lingling ZHANG Na WU Biao GUI Jiangshen
1. School of Information Science and Technology, Zhejiang Sci-Tech University, Hangzhou 310018, China;2. Department of East Asian Studies, Yamaguchi University, Yamaguchi 7538514, Japan
关键词:
人脸检测金字塔网络网络加速小型化级联网络
分类号:
TP181
文献标志码:
A
摘要:
采用卷积神经网络可有效提高人脸检测算法的精度,然而其模型参数过于复杂,在一般设备上检测速度很慢。针对这个问题,提出了一种三层网络级联的人脸检测算法,利用级联方式实现网络小型化,通过多任务方式提高人脸检测的精度。在网络的第一级采用金字塔结构网络,结合anchor机制提取多尺度人脸建议框,在此基础上结合卷积分解策略和网络加速的方法,进一步增强网络特征提取的有效性并减少模型参数。实验结果表明:在FDDB上该算法的检测精度和检测速度均优于MTCNN;在主频为20 GHz的八核设备上,检测速度可以达到80 fps。

参考文献/References:

[1] 梁路宏, 海舟. 脸检测研究综述[J]. 计算机学报, 2002, 25(5):449-458.
[2] Viola P, Jones M J. Robust realtime face detection[J]. International Journal of Computer Vision, 2004,57(2):137-154.
[3] Ojala T, Pietikinen M, Menp T. Multiresolution grayscale and rotation invariant texture classification with local binary patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002,24(7):971-987.
[4] Dalal N, Triggs B. Histograms of oriented gradients for human detection[C]// IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2005. IEEE, 2005, 1: 886-893.
[5] Ng P C, Henikoff S. SIFT: Predicting amino acid changes that affect protein function[J]. Nucleic acids research, 2003, 31(13): 3812-3814.
[6] Wu B, Haizhou A I, Huang C, et al. Fast rotation invariant multiview face detection based on real adaboost[C] //IEEE International Conference on Automatic Face and Gesture Recognition. IEEE, 2004:79-84.
[7] Benini L, Bogliolo A, De Micheli G. A survey of design techniques for systemlevel dynamic power management[J]. IEEE Transactions on Very Large Scale Integration (VLSI) Systems, 2000, 8(3): 299-316.
[8] 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.
[9] Ranjan R, Patel V M, Chellappa R. Hyperface: A deep multitask learning framework for face detection, landmark localization, pose estimation, and gender recognition[J/OL]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017. (2017-12-08) [2018-08-12]. https://doi.org/101109/TPAMI.2017-2781233.
[10] Ren S, He K, Girshick R, et al. Faster rcnn: Towards realtime object detection with region proposal networks[J]. International Conference on Neural Information Processing Systems, 2015, 39(6): 91-99.

备注/Memo

备注/Memo:
作者简介:包晓安(1973-),男,浙江东阳人,教授,硕士,主要从事软件测试、智能信息处理方面的研究
通信作者:张娜,E-mail:zhangna@zstu.edu.cn
更新日期/Last Update: 2019-05-29