REDACTOR OF 3D FACE IMAGE CREATED USING KINECT 2.0 TECHNOLOGY
Abstract
In this paper, we present the results of creating a software product that provides the ability to build 3D image of the face and allows you to solve some of the problems that arise when creating them. To implement the program, the .NET Framework, WPF, Kinect 2.0 SDK, FaceAPI were used.
Some situations situations and problems that arise when creating such an image are considered, as well as possible means and methods of their solution. The possibility of choosing a 3D model that is best suited for the face type, the image of which will be constructed, as well as the optimal fitting of the original 2D image of this face to the selected 3D model was analyzed.
The role of factors influencing the correctness of the process of fitting the two-dimensional image of a person's face to its three-dimensional model was analyzed. For such fit and evaluation of its quality, a method based on the use of “stretching coefficients” was used. The correct choice of values of these coefficients that allow you to compress or stretch an entire image or its specific areas by the amount necessary for the most adequate fit of it to the selected three-dimensional model, makes possible to realize the optimal attachment of the face image to the selected 3D model.
Using the developed program, the process of creating a 3D image of the face was demonstrated, based on its two-dimensional image. It was shown that the main problems that arise in the process of creating a three-dimensional image of a person’s face are the choice of the optimal 3D model of the face, the correct fitting of the two-dimensional image to the selected model of the face and its overlay on this model. These problems can be solved by applying an algorithm using "tensile coefficients" that minimizes distortion of the overlay of a 2D image of a face on a 3D face model and optimizes the overlay process itself. Using this algorithm and the use of calculated fit errors, the selection process of the 3D model of the face, which is best suited for the selected two-dimensional face of the face, and the achievement of minimal distortions when applied to this model, was demonstrated.
The developed program is intended for work in the Windows operating system. To work correctly, you need to use .NET Framework 4.5 or later.
Keywords: 3D face image; 3D face model; face recognition; fitting error; Kinect 2.0; Software Development Kit Kinect.
Full Text:
PDF (Українська)References
Chellappa R. Human and Machine Recognition of Faces / R. Chellappa, C. Wilson, and S. Si-rohey // A Survey. Proc. IEEE, – 1995. – V. 83, no. 5. – p. 705-740.
Tolba A. S. Face Recognition: A Literature Review / A. S. Tolba, A.H. El-Baz, and A.A. El-Harby // International Journal of Signal Processing. – 2006. – V. 2; 2. – p. 88-103.
Belhumeur, J. P. Hespanha. Eigenfaces vs. fisherfaces - recognition using class specific linear projection / J. P. Hespanha Belhumeur, , and D. J. Kriegm // IEEE Transactions on Pattern Analysis and Machine Intelligence – 1997. – V. 19(7). – p. 711–720.
AL-Allaf O. N. A. Review of face detection systems based artificial neural networks algorithms / O. N. A. AL-Allaf // The International Journal of Multimedia & Its Applications. – 2014. – Vol.6, No.1. – p. 1-16.
Tiwari Shradha. A Review of Advancements in Biometric Systems / Shradha Tiwari , J.N. Chourasia, Vijay S.Chourasia // Int. J. of Innovative Research in Advanced Engineering (IJIRAE). – 2015. – V. 2 Is. 1/– p. 187-204.
Smeets Dirk. Objective 3D face recognition: Evolution, approaches and challenges / Dirk Smeets, Peter Claes, Dirk Vandermeulen, John Gerald Clemen // Forensic Science International. – 2010. – 201 – p. 125–132
Bledsoe W. W. The model method in facial recognition. / W.W. Bledsoe – Technical Report PRI 15. – Panoramic Research, Inc. Palo Alto, California. – 1964.
Handbook of Face Recognition. / Eds. S.Z. Li, A.K. Jain – Springer-Verlag, New York. –2005. – p. 398.
Chang K. I. Face recognition using 2D and 3D facial data / K.I. Chang, K.W. Bowyer, P.J. Flynn // ACM Workshop on Multimodal User Authentication. – 2003. p 25-32.
Chellapa R. Human and machine recognition of faces: a survey / R. Chellapa, C.L. Wilson, S. Sirohey. // Proc. IEEE. – 1995. – 83. – p. 705–740.
Zhao W. Face recognition: a literature survey / W. Zhao, R. Chellapa, P.J. Phillips, A. Rosen-feld. // ACM Comput. Surv. – 2003. – 35. – p. 99–458.
Bronstein A. M. Three-Dimensional Face Recognition / A. M. Bronstein, M. M. Bronstein & R. Kimmel // International Journal of Computer Vision. – 2005. – 64(1). – p. 5–30.
Hu Guosheng. Efficient 3D morphable face model fitting / Guosheng Hu , Fei Yan, Josef Kit-tler, William Christmas, Chi Ho Chan, Zhenhua Feng , Patrik Huber // Pattern Recognition. – 2017. – V. 67 – p. 366-379.
Ter Haar F. 3D face model fitting for recognition / F. ter Haar, R. Veltkamp // European Con-ference on Computer Vision. – 2008. – p. 652-664.
Brunton A. Review of statistical shape spaces for 3D data with comparative analysis for human faces / A. Brunton, A. Salazar, T. Bolkart, S. Wuhrer // Comput. Vis. Image Underst. – 2014. – 128 – p. 1–17.
Bas A. Fitting a 3D Morphable Model to Edges: A Comparison Between Hard and Soft Corre-spondences / A. Bas, W. A. P. Smith, T. Bolkart, S. Wuhrer. // Asian Conference on Computer Vision Workshop on Facial Informatics, Nov. 2016, Taipei, Taiwan. Springer, Lecture Notes in Computer Science. – 2017. – vol. 10117. – p. 377-391.
Sela M. Unrestricted Facial Geometry Reconstruction Using Image-To-Image Translation / Matan Sela, Elad Richardson, Ron Kimmel. // The IEEE International Conference on Computer Vision (ICCV). – 2017. – p. 1576-1585.
Zhang Z. Microsoft Kinect sensor and its effect / Z. Zhang // IEEE MultiMedia. – 2012. – V. 19, no. 2. – p. 4–10.
Rahman M. Beginning Microsoft Kinect for Windows SDK 2.0 : Motion and Depth Sensing for Natural Interfaces / M. Rahman. – Apress. – 2017. – 297 p.
Kinect for Windows Software Development Kit (SDK) [Electronic source]. – Available from: https://www.techspot.com/drivers/driver/file/information/18015/.
“WPF overview” [Electronic source]. – Avaible from: https://www.tutorialspoint.com/wpf/wpf_overview.htm
Face Detection using Haar Cascades [Electronic source]. – Available from: https://docs.opencv.org/3.4.1/d7/d8b/tutorial_py_face_detection.html
DOI: http://dx.doi.org/10.30970/eli.14.3
Refbacks
- There are currently no refbacks.