[1]吕 晶,薛亚非,谷静平.像素邻域差向量协方差特征的多视角人脸检测[J].南京师范大学学报(工程技术版),2022,22(02):073-79.[doi:10.3969/j.issn.1672-1292.2022.02.011]
 Lü Jing,Xue Yafei,Gu Jingping.Multi-view Face Detection Based on the Pixel Differential Covariance Feature[J].Journal of Nanjing Normal University(Engineering and Technology),2022,22(02):073-79.[doi:10.3969/j.issn.1672-1292.2022.02.011]
点击复制

像素邻域差向量协方差特征的多视角人脸检测
分享到:

南京师范大学学报(工程技术版)[ISSN:1006-6977/CN:61-1281/TN]

卷:
22卷
期数:
2022年02期
页码:
073-79
栏目:
计算机科学与技术
出版日期:
2022-06-30

文章信息/Info

Title:
Multi-view Face Detection Based on the Pixel Differential Covariance Feature
文章编号:
1672-1292(2022)02-0073-07
作者:
吕 晶薛亚非谷静平
(南京师范大学中北学院,江苏 丹阳 212300)
Author(s):
Lü JingXue YafeiGu Jingping
(Zhongbei College,Nanjing Normal University,Danyang 212300,China)
关键词:
人脸检测像素差向量特征像素差向量协方差特征人脸候选区域提取
Keywords:
face detectionpixel differential featurepixel differential covariance featureface proposal
分类号:
TP391.41
DOI:
10.3969/j.issn.1672-1292.2022.02.011
文献标志码:
A
摘要:
像素差向量特征是一阶统计量,而研究其高阶统计特性仍是一个未解问题. 针对该问题,研究了像素差向量在局部邻域中的协方差统计特性,提出了一种新颖的像素局部差向量协方差特征,显著提高了多视角人脸检测的性能. 该方法首先计算输入图像的多通道特征图,并计算特征图的像素差向量; 然后在指定尺寸的邻域内计算像素差向量的协方差矩阵,获得差向量中任意两个元素间的相关信息,用于表示人脸的局部特征; 最后设计了一种人脸候选区域提取方法,对该人脸检测器作进一步优化,实现了实时运行速度. 实验结果表明,本文方法优于当前其他主流方法,且可部署在低功耗的边缘计算设备上.
Abstract:
PDV is a discriminative first order operator,how to encode high order statistics into PDV is still unaddressed. In this paper,we investigate the covariance of the PDV in a local region,and propose a novel pixel differential covariance feature(PDCF),significantly improving the performance for multi-view face detection. Following the PDV based object detection pipeline,multiple channel maps are initially prepared before calculating pixel differential feature. Afterwards,covariance matrix of the PDV is computed to obtain the pair-wise correlation between arbitrary two elements. Finally,a face proposal mechanism is designed to generate face candidates and a PDCF based detector is further refined,achieving real-time running speed. Experimental results show that,the proposed method outperforms other state-of-the-art methods and can be deployed on the edge-computing device.

参考文献/References:

[1] 曾建凡. 多角度人脸检测与识别方法研究[J]. 电子设计工程,2017,25(11):41-44.
[2]YANG B,YAN J J,LEI Z,et al. Aggregate channel features for multi-view face detection[C]//Proceedings of the 2014 International Joint Conference on Biometrics. Clearwater,USA:IEEE,2014.
[3]LI J G,ZHANG Y M. Learning SURF cascade for fast and accurate object detection[C]//Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition. Portland,USA:IEEE,2013.
[4]SHEN J F,ZUO X,LI J,et al. A novel pixel neighborhood differential statistic feature for pedestrian and face detection[J]. Pattern Recognition,2017,63:127-138.
[5]张明浩,杨耀权,靳渤文. 基于图像增强技术的SURF特征匹配算法研究[J]. 自动化与仪表,2019,34(9):98-102.
[6]UEHARA K,SAKANASHI H,NOSATO H,et al. Object detection of satellite images using multi-channel higher-order local autocorrelation[C]//2017 IEEE International Conference on Systems,Man,and Cybernetics. Banff,Canada:IEEE,2017.
[7]MATHIAS M,PEDERSOLI R B M,VAN R B M. Gool,face detection without bells and whistles[C]//Proceedings of the 13th European Conference on Computer Vision. Zurich,Switzerland:Springer,2014.
[8]耿渊哲. 基于LBP采样学习的人脸识别研究[J]. 计算机与数字工程,2018,46(2):371-374.
[9]YANG S,LUO P,LOY C C,et al. Faceness-Net:face detection through deep facial part responses[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2018,40(8):1845-1859.
[10]CHEN D,HUA G,WEN F,et al. Supervised transformer network for efficient face detection[C]//Proceedings of the 14th European Conference on Computer Vision. Amsterdan,Netherlands:Springer,2016.
[11]LI J,WANG Y B,WANG C A,et al. DSFD:dual shot face detector[C]//Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach,USA:IEEE,2019.
[12]ZHANG S,CHI C,LEI Z,et al. RefineFace:refinement neural network for high performance face detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,43(11):4008-4020.

相似文献/References:

[1]谢 非,龚 俊,王元祥,等.基于肤色增强和分块PCA的人脸表情识别方法[J].南京师范大学学报(工程技术版),2017,17(02):049.[doi:10.3969/j.issn.1672-1292.2017.02.008]
 Xie Fei,Gong Jun,Wang Yuanxiang,et al.A Facial Expression Recognition Method Based onSkin Color Enhancement and Block PCA[J].Journal of Nanjing Normal University(Engineering and Technology),2017,17(02):049.[doi:10.3969/j.issn.1672-1292.2017.02.008]
[2]曹金梦,倪蓉蓉,杨 彪.面向面部表情识别的双通道卷积神经网络[J].南京师范大学学报(工程技术版),2018,18(03):001.[doi:10.3969/j.issn.1672-1292.2018.03.001]
 Cao Jinmeng,Ni Rongrong,Yang Biao.Binary-Channel Convolutional Neural Network forFacial Expression Recognition[J].Journal of Nanjing Normal University(Engineering and Technology),2018,18(02):001.[doi:10.3969/j.issn.1672-1292.2018.03.001]

备注/Memo

备注/Memo:
收稿日期:2021-06-24.
基金项目:江苏省高等学校自然科学基金项目(19KJB52004).
通讯作者:吕晶,讲师,研究方向:图像处理. E-mail:51818715@qq.com
更新日期/Last Update: 1900-01-01