基于RGB-D图像的头部姿态检测

计算机科学 ›› 2019, Vol. 46 ›› Issue (11A): 334-340.

基于RGB-D图像的头部姿态检测

刘振宇, 关彤

(沈阳工业大学信息科学与工程学院沈阳110870)

出版日期:2019-11-10 发布日期:2019-11-20
通讯作者: 关彤(1994-),女,硕士生,主要研究方向为视觉伺服、图像处理,E-mail:18842414590@163.com。
作者简介:刘振宇(1973-),男,博士,教授,主要研究方向为视觉伺服、模式识别。
基金资助:
本文受辽宁省2018年自然科学基金(20180520022)资助。

Head Posture Detection Based on RGB-D Image

LIU Zhen-yu, GUAN Tong

(School of Information Science and Engineering,Shenyang University of Technology,Shenyang 110870,China)

Online:2019-11-10 Published:2019-11-20

摘要/Abstract

摘要： 在经颅磁刺激治疗过程中,准确且快速地检测人体头部姿态至关重要。针对基于二维彩色图像的头部姿态估计对环境、姿态敏感的问题,文中提出了一种联合彩色图像与深度图像的头部姿态检测方法。通过彩色图像检测人脸特征点的二维位置信息,结合深度信息定义三维头部坐标系;然后在现有的ICP点云配准算法的基础上,提出了一种粗配准方法。通过计算待检测头部点云与标准头部点云之间坐标系的变换关系得到初始位姿参数,以防止点云配准陷入局部最优局面。实验表明,该算法能够在光源均匀且充足的诊疗室环境中准确地检测人体头部姿态,提高头部姿态角度大时姿态估计的鲁棒性。

关键词: RGB-D图像, 点云配准, 人脸特征点, 三维头部坐标系, 头部姿态检测

Abstract: In the process of transcranial magnetic stimulation treatment,it is important to accurately and quickly detect the posture of the human head.Aiming at the problem that the head pose estimation based on two-dimensional color image is sensitive to the environment and posture,a head posture detection method combining both color image and depth image was proposed.The two-dimensional position information of the face feature points is detected by the color image,and the three-dimensional head coordinate system is defined by combining the depth information;Then,based on the existing ICP point cloud registration algorithm,a coarse registration method was proposed.The initial pose parameters are obtained by calculating the transformation relationship between the coordinate system of the head cloud,to be detected and the standard head point cloud,to protect the point cloud registration from falling into local optimum.Experiments show that the algorithm can accurately detect the head posture of the human body in a consulting room environment where the light source is uniform and sufficient,and improve the robustness of the attitude estimation when the head posture angle is large.

Key words: Face feature points, Head posture detection, Point cloud registration, RGB-D image, Three-dimensional head coordinate system

中图分类号:

TP391.41

刘振宇, 关彤. 基于RGB-D图像的头部姿态检测[J]. 计算机科学, 2019, 46(11A): 334-340. https://doi.org/

LIU Zhen-yu, GUAN Tong. Head Posture Detection Based on RGB-D Image[J]. Computer Science, 2019, 46(11A): 334-340. https://doi.org/

参考文献

[1]MURPHY-CHUTORIAN E.Head Pose Estimation in ComputerVision:A Survey [J].IEEE Transactions on Pattern Analysis &Machine Intelligence,2009,31(4):607-626.
[2]MEYER G P,GUPTA S,FROSIO I,et al.Robust Model-Based 3D Head Pose Estimation [C]∥IEEE International Conference on Computer Vision.IEEE,2016.
[3]MORENCY L P.3D Constrained Local Model for rigid and non-rigid facial tracking [C]∥Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE,2012.
[4]PAPAZOV C,MARKS T K,JONES M.Real-time 3D head pose and facial landmark estimation from depth images using triangular surface patch features [C]∥Computer Vision & Pattern Recognition.IEEE,2015.
[5]TAN D J,TOMBARI F,NAVAB N.Real-Time Accurate 3D Head Tracking and Pose Estimation with Consumer RGB-D Cameras [J].International Journal of Computer Vision,2018,126(2／3／4):158-183.
[6]FANELLI G.Real Time Head Pose Estimation from Consumer Depth Cameras [C]∥International Conference on Pattern Re-cognition.Springer-Verlag,2011.
[7]郭知智,周前祥,柳忠起.基于自适应线性回归的头部姿态计算[J].计算机应用研究,2016,33(10):3181-3184.
[8]SUN Y,YIN L.Automatic Pose Estimation of 3D Facial Models [C]∥International Conference on Pattern Recognition.IEEE,2008.
[9]PADELERIS P,ZABULIS X,ARGYROS A.Head pose estimation on depth data based on Particle Swarm Optimization [C]∥Workshop on Human Activity Understanding from 3d Data.IEEE,2012.
[10]WANG B,LIANG W,WANG Y,et al.Head pose estimation with combined 2D SIFT and 3D HOG features [C]∥Seventh International Conference on Image & Graphics.IEEE,2013.
[11]YANG J,WEI L,JIA Y.Face pose estimation with combined 2D and 3D HOG features [C]∥International Conference on Pattern Recognition.2012.
[12]MADRIGAL F,LERASLE F,MONIN A.3D Head Pose Estimation Enhanced Through SURF-Based Key-Frames [C]∥IEEE Winter Conference on Applications of Computer Vision.IEEE Computer Society,2018.
[13]唐云祁,孙哲南,谭铁牛.头部姿势估计研究综述[J].模式识别与人工智能,2014,27(3):213-225.
[14]闵秋莎,刘能,陈雅婷,等.基于面部特征点定位的头部姿态估计[J].计算机工程,2018,44(6):263-269.
[15]SEEMANN E,NICKEL K,STIEFELHAGEN R.Head pose estimation using stereo vision for human-robot interaction [C]∥IEEE International Conference on Automatic Face & Gesture Recognition.IEEE Computer Society,2004.
[16]DERKACH D,RUIZ A,SUKNO F M,et al.Head Pose Estimation Based on 3-D Facial Landmarks Localization and Regression [C]∥IEEE International Conference on Automatic Face & Gesture Recognition.IEEE,2017.
[17]IIRIS L,ESCARELA S,ANBARJAFARI G.SASE:RGB-Depth Database for Human Head Pose Estimation [C]∥European Conference on Computer Vision.Springer International Publishing,2016.
[18]李成龙.基于RGB-D数据的头部姿态估计研究[D].济南:山东大学,2017.
[19]尹婕.基于图像信息的点云优化研究[D].成都:电子科技大学,2017.
[20]ROUTRAY S,RAY A K,MISHRA C.Image denoising by preserving geometric components based on weighted bilateral filter and curvelet transform [J].Optik,2018:S0030402618301153.
[21]张鸿宇,刘威,许炜,等.基于深度图像的多学习者姿态识别[J].计算机科学,2015,42(9):299-302.
[22]郭连朋,陈向宁,刘彬.Kinect传感器的彩色和深度相机标定[J].中国图象图形学报,2014,19(11):1584-1590.
[23]ÇELIKTUTAN O,ULUKAYA S,SANKUR B.A comparative study of face landmarking techniques [J].Eurasip Journal on Image & Video Processing,2013,2013(1):13.
[24]MARACI M A,NAPOLITANO R,PAPAGEORGHIOU A,et al.Object classification in an ultrasound video using LP-SIFT features [M]∥Medical Computer Vision:Algorithms for Big Data.Springer International Publishing,2014.
[25]KAZEMI V,SULLIVAN J.One Millisecond Face Alignmentwith an Ensemble of Regression Trees [C]∥2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).IEEE Computer Society,2014.
[26]蔡自兴,谢斌.机器人学[M].北京:清华大学出版社,2015:20.
[27]李良福,邹彬,周国良,等.基于优化估计的深度图像修复与误差补偿方法研究[J].应用光学,2018(1):45-50.
[28]BESL P J,MCKAY N D.A Method for Registration of 3-DShapes [M].IEEE Computer Society,1992.

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed