A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method

Zhong Qiu Zhao; Kewen Cheng; Qinmu Peng; Xindong Wu

doi:10.1007/978-3-319-42291-6_8

A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method

Zhong Qiu Zhao^*, Kewen Cheng, Qinmu Peng, Xindong Wu

^*Corresponding author for this work

Department of Computer Science

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

Abstract

In this paper, we proposed a real-time head pose estimation algorithm by extending Pose from Orthography and Scaling with Iterations (POSIT) (named Adaptive POSIT) method and modifying the Supervised Descent Method (SDM). Specifically, we used the modified SDM for facial landmarks detection and tracking, and adopted adaptive POSIT to estimate head pose. In the feature selection stage, we extracted different features in neighboring facial landmarks instead of a single feature. In the facial landmarks selection stage, we used partial facial landmarks instead of the whole facial landmarks. The experiments show that our method can track facial landmarks robustly with tolerance to certain illumination changes and partial occlusion, and improves the accuracy of head pose estimation.

Original language	English
Title of host publication	Intelligent Computing Theories and Application
Subtitle of host publication	12th International Conference, ICIC 2016, Lanzhou, China, August 2-5, 2016, Proceedings, Part I
Editors	De-Shuang Huang, Vitoantonio Bevilacqua, Prashan Premaratne
Publisher	Springer Cham
Pages	74-85
Number of pages	12
Edition	1st
ISBN (Electronic)	9783319422916
ISBN (Print)	9783319422909
DOIs	https://doi.org/10.1007/978-3-319-42291-6_8
Publication status	Published - 11 Jul 2016
Event	12th International Conference on Intelligent Computing Theories and Application, ICIC 2016 - Lanzhou, China Duration: 2 Aug 2016 → 5 Aug 2016

Publication series

Name	Lecture Notes in Computer Science
Volume	9771
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Name	Information Systems and Applications, incl. Internet/Web, and HCI

Name	ICIC: International Conference on Intelligent Computing

Conference

Conference	12th International Conference on Intelligent Computing Theories and Application, ICIC 2016
Country/Territory	China
City	Lanzhou
Period	2/08/16 → 5/08/16

User-Defined Keywords

Facial landmarks
Head pose estimation
POSIT
SDM

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-3-319-42291-6_8

Cite this

Zhao, Z. Q., Cheng, K., Peng, Q., & Wu, X. (2016). A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method. In D.-S. Huang, V. Bevilacqua, & P. Premaratne (Eds.), Intelligent Computing Theories and Application: 12th International Conference, ICIC 2016, Lanzhou, China, August 2-5, 2016, Proceedings, Part I (1st ed., pp. 74-85). (Lecture Notes in Computer Science; Vol. 9771), (Information Systems and Applications, incl. Internet/Web, and HCI), (ICIC: International Conference on Intelligent Computing). Springer Cham. https://doi.org/10.1007/978-3-319-42291-6_8

Zhao, Zhong Qiu ; Cheng, Kewen ; Peng, Qinmu et al. / A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method. Intelligent Computing Theories and Application: 12th International Conference, ICIC 2016, Lanzhou, China, August 2-5, 2016, Proceedings, Part I. editor / De-Shuang Huang ; Vitoantonio Bevilacqua ; Prashan Premaratne. 1st. ed. Springer Cham, 2016. pp. 74-85 (Lecture Notes in Computer Science). (Information Systems and Applications, incl. Internet/Web, and HCI). (ICIC: International Conference on Intelligent Computing).

@inproceedings{8ae412000fbb4a3c84196984a2d44284,

title = "A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method",

abstract = "In this paper, we proposed a real-time head pose estimation algorithm by extending Pose from Orthography and Scaling with Iterations (POSIT) (named Adaptive POSIT) method and modifying the Supervised Descent Method (SDM). Specifically, we used the modified SDM for facial landmarks detection and tracking, and adopted adaptive POSIT to estimate head pose. In the feature selection stage, we extracted different features in neighboring facial landmarks instead of a single feature. In the facial landmarks selection stage, we used partial facial landmarks instead of the whole facial landmarks. The experiments show that our method can track facial landmarks robustly with tolerance to certain illumination changes and partial occlusion, and improves the accuracy of head pose estimation.",

keywords = "Facial landmarks, Head pose estimation, POSIT, SDM",

author = "Zhao, {Zhong Qiu} and Kewen Cheng and Qinmu Peng and Xindong Wu",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2016.; 12th International Conference on Intelligent Computing Theories and Application, ICIC 2016 ; Conference date: 02-08-2016 Through 05-08-2016",

year = "2016",

month = jul,

day = "11",

doi = "10.1007/978-3-319-42291-6_8",

language = "English",

isbn = "9783319422909",

series = "Lecture Notes in Computer Science",

publisher = "Springer Cham",

pages = "74--85",

editor = "De-Shuang Huang and Vitoantonio Bevilacqua and Prashan Premaratne",

booktitle = "Intelligent Computing Theories and Application",

edition = "1st",

}

Zhao, ZQ, Cheng, K, Peng, Q & Wu, X 2016, A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method. in D-S Huang, V Bevilacqua & P Premaratne (eds), Intelligent Computing Theories and Application: 12th International Conference, ICIC 2016, Lanzhou, China, August 2-5, 2016, Proceedings, Part I. 1st edn, Lecture Notes in Computer Science, vol. 9771, Information Systems and Applications, incl. Internet/Web, and HCI, ICIC: International Conference on Intelligent Computing, Springer Cham, pp. 74-85, 12th International Conference on Intelligent Computing Theories and Application, ICIC 2016, Lanzhou, China, 2/08/16. https://doi.org/10.1007/978-3-319-42291-6_8

A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method. / Zhao, Zhong Qiu; Cheng, Kewen; Peng, Qinmu et al.
Intelligent Computing Theories and Application: 12th International Conference, ICIC 2016, Lanzhou, China, August 2-5, 2016, Proceedings, Part I. ed. / De-Shuang Huang; Vitoantonio Bevilacqua; Prashan Premaratne. 1st. ed. Springer Cham, 2016. p. 74-85 (Lecture Notes in Computer Science; Vol. 9771), (Information Systems and Applications, incl. Internet/Web, and HCI), (ICIC: International Conference on Intelligent Computing).

Research output: Chapter in book/report/conference proceeding › Conference proceeding › peer-review

TY - GEN

T1 - A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method

AU - Zhao, Zhong Qiu

AU - Cheng, Kewen

AU - Peng, Qinmu

AU - Wu, Xindong

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2016.

PY - 2016/7/11

Y1 - 2016/7/11

N2 - In this paper, we proposed a real-time head pose estimation algorithm by extending Pose from Orthography and Scaling with Iterations (POSIT) (named Adaptive POSIT) method and modifying the Supervised Descent Method (SDM). Specifically, we used the modified SDM for facial landmarks detection and tracking, and adopted adaptive POSIT to estimate head pose. In the feature selection stage, we extracted different features in neighboring facial landmarks instead of a single feature. In the facial landmarks selection stage, we used partial facial landmarks instead of the whole facial landmarks. The experiments show that our method can track facial landmarks robustly with tolerance to certain illumination changes and partial occlusion, and improves the accuracy of head pose estimation.

AB - In this paper, we proposed a real-time head pose estimation algorithm by extending Pose from Orthography and Scaling with Iterations (POSIT) (named Adaptive POSIT) method and modifying the Supervised Descent Method (SDM). Specifically, we used the modified SDM for facial landmarks detection and tracking, and adopted adaptive POSIT to estimate head pose. In the feature selection stage, we extracted different features in neighboring facial landmarks instead of a single feature. In the facial landmarks selection stage, we used partial facial landmarks instead of the whole facial landmarks. The experiments show that our method can track facial landmarks robustly with tolerance to certain illumination changes and partial occlusion, and improves the accuracy of head pose estimation.

KW - Facial landmarks

KW - Head pose estimation

KW - POSIT

KW - SDM

UR - http://www.scopus.com/inward/record.url?scp=84978792028&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-42291-6_8

DO - 10.1007/978-3-319-42291-6_8

M3 - Conference proceeding

AN - SCOPUS:84978792028

SN - 9783319422909

T3 - Lecture Notes in Computer Science

SP - 74

EP - 85

BT - Intelligent Computing Theories and Application

A2 - Huang, De-Shuang

A2 - Bevilacqua, Vitoantonio

A2 - Premaratne, Prashan

PB - Springer Cham

T2 - 12th International Conference on Intelligent Computing Theories and Application, ICIC 2016

Y2 - 2 August 2016 through 5 August 2016

ER -

Zhao ZQ, Cheng K, Peng Q, Wu X. A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method. In Huang DS, Bevilacqua V, Premaratne P, editors, Intelligent Computing Theories and Application: 12th International Conference, ICIC 2016, Lanzhou, China, August 2-5, 2016, Proceedings, Part I. 1st ed. Springer Cham. 2016. p. 74-85. (Lecture Notes in Computer Science). (Information Systems and Applications, incl. Internet/Web, and HCI). (ICIC: International Conference on Intelligent Computing). doi: 10.1007/978-3-319-42291-6_8

A Real-Time Head Pose Estimation Using Adaptive POSIT Based on Modified Supervised Descent Method

Abstract

Publication series

Conference

User-Defined Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this