3D face detection based on salient features extraction and skin colour detection using data mining

ABSTRACT

Face detection has an essential role in many applications. In this paper, we propose an efficient and robust method for face detection on a 3D point cloud represented by a weighted graph. This method classifies graph vertices as skin and non-skin regions based on a data mining predictive model. Then, the saliency degree of vertices is computed to identify the possible candidate face features. Finally, the matching between non-skin regions representing eyes, mouth and eyebrows and salient regions is done by detecting collisions between polytopes, representing these two regions. This method extracts faces from situations where pose variation and change of expressions can be found. The robustness is showed through different experimental results. Moreover, we study the stability of our method according to noise. Furthermore, we show that our method deals with 2D images.

KEYWORDS:

• 3D facial point clouds detection
• 3D point clouds saliency
• 3D point clouds segmentation
• classification
• 3D object processing
• points clouds

Notes

Please see below for information relating to all images reproduced throughout this article:

Figures 4, 11 and 12. 3D Objects which can be obtained from https://sketchfab.com/models

Figure 15. 3D meshes created from 2D images using specific tools. Original images obtained from the following repositories:

1. http://www.german-pathway-colleges.org/wp-content/uploads/2015/11/International_students-300x266.jpg

2. https://www.ronitbaras.com/family-matters/parenting-family/three-is-a-crowd/

3. http://www.bytedigital.co.uk/who-we-are/our-company/

Figure 18. © The Authors. Reproduced with permission.

Figures 19 and 20. Used from dataset of Reference [2] (Moudani W, Sayed AR, 2010), also found at http://www.aiai.ed.ac.uk/project/plan/peapod/2002-05-13.html

Figure 21. Image obtained from https://ca.news.yahoo.com/roadside-body-could-missing-big-man-190107548.html

Figures 22 and 23. Used from experimental results of Reference [3] (Bin Ghazali KH, Ma J, Xiao R, 2011) in order to gain comparison.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Abdul Rahman El Sayed received the Master I degree in applied mathematics, from the Lebanese University, in 2002, the Master II degree (with honours) in informatics, from the Lebanese University, in 2008 and the M. Eng. in computer engineering from the University of CNAM, Beirut, Lebanon, Center affiliated to CNAM Paris, in 2015. He has held lecturing positions at Tripoli Technical Institute and Tripoli Industrial Institute, Lebanon. In addition, he is an IT manager at Tripoli Governmental Hospital and the owner of a programming company in North of Lebanon. His research interests cover the image processing 2D and 3D with many publications in this scope.

Abdallah El Chakik received his Ph.D. and M.S. in computer science, both from the University of Caen, France. He received his B.S. in computer science from the University of Saint Etienne-France in 2006. He is an assistant professor in the Department of Computer Science at Beirut Arab University. He worked as a research engineer in Euridis Ingénierie, France.

Hassan Alabboud has been an assistant professor in the Lebanese University in Tripoli, Lebanon since 2008. He received his Ph.D. in 2007 at University of Le Havre in Le Havre, France. His Ph.D. thesis developed several methods and approaches in numerical optimization and operations research for binary and continuous Quadratic Programming Problems. He obtained the Master II degree in Pure Mathematics from the University of Sciences and Technology of Lille, France in 2003 and the Master I degree in Pure Mathematics from the Lebanese University in Tripoli, Lebanon in 2002. His research interests are combinatorial optimization, DC optimization, semidefinite programming, quadratic programming, crisis management and Image processing.

Adnan Yassine is a full professor in the University of Le Havre to Normandy in France and a member of the Laboratory of Applied Mathematics of Le Havre (LMAH). He teaches optimization and logistics in the Superior Institute of Logistic Studies (ISEL). He is author and co-author of numerous international publications in numerical and combinatorial optimization some of which concern the scheduling and logistics problems. He has supervised some thirty Ph.Ds. He obtained a Ph.D. in Applied Mathematics from the University of Grenoble (France) in 1989 and Habilitation to Supervise Research (HDR) from the University of Nancy (France) in 1998.