一种鲁棒的人脸关键点实时跟踪方法

doi:10.3969/j.issn.1000-3428.2018.04.045

摘要/Abstract

摘要： 针对视频图像序列中人脸关键点跟踪对鲁棒性和实时运行的要求,提出一种新的人脸关键点实时跟踪方法。运用光流法跟踪若干显著关键点,为下一帧选择更好的初始形状,根据当前帧的人脸形状估计下一帧的人脸框,以减少对人脸检测器的依赖,同时为防止误差累积,加入人脸检测器重启机制。实验结果表明,该方法在300-VW数据集上实现了68个人脸关键点的鲁棒跟踪,运行速度达30+f/s,可用于大多数人脸相关的实时应用。

关键词: 人脸关键点, 局部二值特征, 显著关键点, 初始形状, 光流

Abstract: Aiming at the requirement of robustness and real-time running of facial key point tracking in video sequence,this paper proposes a simple and effective real-time tracking method of facial key points.The optical flow method is used to track a number of significant points and select a better initial shape for the next frame.The face frame of the next frame is estimated according to the face shape of the current frame to reduce the dependence on the face detector.To prevent errors accumulation,the face detector restart mechanism is joined.Experimental results on a 300-VW dataset show that this method can achieve robust tracking of 68 face key points with a speed of 30+f/s,which can be used for most face real-time applications.

Key words: facial key point, Local Binary Feature(LBF), significant key point, initial shape, optical flow

中图分类号:

TP391

徐威威,李俊. 一种鲁棒的人脸关键点实时跟踪方法[J]. 计算机工程, doi: 10.3969/j.issn.1000-3428.2018.04.045.

XU Weiwei,LI Jun. A Robust Real-time Tracking Method of Facial Key Point[J]. Computer Engineering, doi: 10.3969/j.issn.1000-3428.2018.04.045.

https://www.ecice06.com/CN/Y2018/V44/I4/281

参考文献

参考文献［1］WAGNER A,WRIGHT J,GANESH A,et al.Toward a practical face recognition system:robust alignment and illumination by sparse representation［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(2):372-386. ［2］ZENG Z,PANTIC M,ROISMAN G I,et al.A Survey of affect recognition methods:audio,visual,and spontaneous expressions［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(1):39-58. ［3］MURPHY C E,TRIVEDI M M.Head pose estimation in computer vision:a survey［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2009,31(4):607-626. ［4］CHRYSOS G G,ANTONAKOS E,ZAFEIRIOU S,et al.Offline deformable face tracking in arbitrary videos［C］//Proceedings of IEEE International Conference on Computer Vision Workshops.Washington D.C.,USA:IEEE Press,2015:1-9. ［5］SHEN J,ZAFEIRIOU S,CHRSOS G G,et al.The first facial landmark tracking in-the-wild challenge:benchmark and results［C］//Proceedings of IEEE International Con-ference on Computer Vision Workshop.Washington D.C.,USA:IEEE Press,2015:1003-1011. ［6］TZIMIROPOULOS G.Project-out cascaded regression with an application to face alignment［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:3659-3667. ［7］COOTES T F,EDWARDS G J,TAYLOR C J.Active appearance models［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence,2001,23(6):681-685. ［8］CRISTINACCE D,COOTES T F.Feature detection and tracking with constrained local models［C］//Proceedings of BMVC’06.London,UK:［s.n.］,2006:929-938. ［9］DOLLAR P,WELINDER P,PERONA P.Cascaded pose regression［C］//Proceedings of 2010 IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2010:1078-1085. ［10］PENG X,FERIS R S,WANG X,et al.A recurrent encoder-decoder network for sequential face alignment［C］//Proceedings of European Conference on Computer Vision.Berlin,Germany:Springer,2016:38-56. ［11］黄琛,丁晓青,方驰.一种鲁棒高效的人脸特征点跟踪方法［J］.自动化学报,2012,38(5):788-796. ［12］郭修宵,陈莹.非约束环境下人脸特征点的稳定跟踪［J］.计算机辅助设计与图形学学报,2014,26(7):1135-1142. ［13］URICAR M,FRANC V,HLAVAC V.Facial landmark tracking by tree-based deformable part model based detector［C］//Proceedings of IEEE International Conference on Computer Vision Workshops.Washington D.C.,USA:IEEE Press,2015:10-17. ［14］WU Y,JI Q.Shape augmented regression method for face alignment［C］//Proceedings of IEEE International Conference on Computer Vision Workshops.Washington D.C.,USA:IEEE Press,2015:26-32. ［15］YANG J,DENG J,ZHANG K,et al.Facial shape tracking via spatio-temporal cascade shape regression［C］//Proceedings of IEEE International Conference on Computer Vision Workshops.Washington D.C.,USA:IEEE Press,2015:41-49. ［16］XIAO Shengtao,YAN Shuicheng,Kassim A A.Facial landmark detection via progressive initialization［C］//Proceedings of IEEE International Conference on Computer Vision Workshops.Washington D.C.,USA:IEEE Press,2015:33-40. ［17］REN S,CAO X,WEI Y,et al.Face alignment at 3 000 f/s via regressing local binary features［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2014:1685-1692. ［18］CAO X,WEI Y,WEN F,et al.Face alignment by explicit shape regression［J］.International Journal of Computer Vision,2014,107(2):177-190. ［19］BREIMAN L.Random forests［J］.Machine Learning,2001,45(1):5-32. ［20］BURGOS-ARTIZZU X P,PERONA P,DOLLAR P.Robust face landmark estimation under occlusion［C］//Proceedings of IEEE International Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2013:1513-1520. ［21］XIONG X.Supervised descent method and its applications to face alignment［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2013:532-539. ［22］BOUGUET J Y.Pyramidal implementation of the Lucas Kanade feature tracker description of the algorithm［J］.Opencv Documents,1999,22(2):363-381. ［23］Libfacedetection［EB/OL］.［2014-02-15］.https://github.com/ShiqiYu/libfacedetection. ［24］ASTHANA A,ZAFEIRIOU S,CHENG S,et al.Incremental face alignment in the wild［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2014:1859-1866. 编辑索书志

[1]	杨硕, 王一丁. 基于改进薄板样条运动模型的人脸动画算法[J]. 计算机工程, 2024, 50(6): 255-265.
[2]	周秦源, 邓越平, 张磊, 张陈, 卢日荣, 胡贤哲. 融合光流与多视角几何的动态视觉SLAM系统[J]. 计算机工程, 2024, 50(5): 250-259.
[3]	杜田田, 王晓龙, 何劲. 复杂光照条件下基于光流的水运航道流速检测算法[J]. 计算机工程, 2024, 50(4): 60-67.
[4]	徐春波, 闫娟, 杨慧斌, 王博, 吴晗. 基于目标检测和语义分割的视觉SLAM算法[J]. 计算机工程, 2023, 49(8): 199-206, 214.
[5]	伍子嘉, 陈航, 彭勇, 宋威. 动态环境下融合轻量级YOLOv5s的视觉SLAM[J]. 计算机工程, 2022, 48(8): 187-195,205.
[6]	李婕, 周顺. 基于超像素和光流引导的正射影像拼接方法[J]. 计算机工程, 2022, 48(3): 263-270.
[7]	梁正友, 刘德志, 孙宇. 结合迁移学习与可分离三维卷积的微表情识别方法[J]. 计算机工程, 2022, 48(1): 228-235.
[8]	季皓宣烨, 梁鹏鹏, 柴玉梅, 王黎明. 基于关键点及光流的平面物体跟踪算法[J]. 计算机工程, 2021, 47(4): 234-240.
[9]	郑伟成, 李学伟, 刘宏哲, 代松银. 基于深度学习的疲劳驾驶检测算法[J]. 计算机工程, 2020, 46(7): 21-29.
[10]	罗凡波, 王平, 梁思源, 徐桂菲, 王伟. 基于深度学习与稀疏光流的人群异常行为识别[J]. 计算机工程, 2020, 46(4): 287-293,300.
[11]	胡斐,文畅,谢凯,贺建飚. 基于微调策略的多线索融合人脸活体检测[J]. 计算机工程, 2019, 45(5): 256-260.
[12]	马钰锡,谭励,董旭,于重重. 面向VTM的交互式活体检测算法[J]. 计算机工程, 2019, 45(3): 256-261.
[13]	金泽芬芬,侯志强,余旺盛,王鑫. 基于博弈论的颜色与运动特征融合跟踪[J]. 计算机工程, 2018, 44(8): 237-242.
[14]	张文达,许悦雷,马时平,李帅,邹洪中. 基于多尺度V1-MT前馈模型的光流计算方法[J]. 计算机工程, 2017, 43(9): 205-209.
[15]	熊饶饶,胡学敏,陈龙,周慧子. 利用综合光流直方图的人群异常行为检测[J]. 计算机工程, 2017, 43(10): 228-233.

选择文件类型/文献管理软件名称

选择包含的内容