Abstract
3D face reconstruction from multi-view video sequences has become a hotspot in computer vision for the last decades. Structure from Motion (SfM) methods, which have been widely used for multi-view 3D face reconstruction, have two main limitations. First, self-occlusion causes certain facial feature points (FFPs) to be invisible in the images, which will lead to missing data. The existing SfM methods could recover the missing data through iterative calculation, however, with high computational costs and long processing time. Second, the SfM methods cannot reconstruct the accurate 3D facial shapes of cheeks because there are no FFPs in this area. This paper proposes a novel “coarse-to-fine” multi-view 3D face reconstruction method by taking the advantage of the complementarity between FFPs and occluding contours, i.e., the boundary lines depicted between the facial region and the background. In this method, a block SfM algorithm is firstly proposed to reconstruct a “coarse” 3D facial shape by utilizing sparse FFPs. The block SfM algorithm does not estimate the true locations of the self-occluded FFPs iteratively. Thus, the computational cost is significantly reduced. Then, a kernel partial least squares (KPLS) algorithm is introduced to refine the “coarse” 3D facial shape. The KPLS method applies occluding contours to remedy the limitation of sparse FFPs correspondence-based SfM method. The proposed method is evaluated on the synthetic sequences generated from the BJUT-3D face database and the real-world multi-view video sequences obtained in a controlled indoor environment. The results show improvements in both accuracy and efficiency.
Similar content being viewed by others
References
Amberg B, Blake A, Fitzgibbon A, et al (2007) Reconstructing high quality face-surfaces using model based stereo. IEEE 11th International Conference on Computer Vision (ICCV) 2007:1–8
Aissaoui A, Martinet J, Djeraba C (2014) Rapid and accurate face depth estimation in passive stereo systems. Multimedia Tools and Applications 72(3):2413–2438
Angelopoulou M, Petrou M (2014) Uncalibrated flatfielding and illumination vector estimationfor photometric stereo face reconstruction. Mach Vis Appl 25(5):1317–1332
Bookstein FL (1989) Principal warps: Thin-plate splines and the decomposition of deformations. IEEE Trans Pattern Anal Mach Intell 6:567–585
Bookstein Fred L (1997) Morphometric tools for landmark data: geometry and biology. Cambridge University Press
Blanz V, Vetter T (1999) A morphable model for the synthesis of 3D faces. Proceedings of the 26th annual conference on Computer graphics and interactive techniques 1999:187–194
Chowdhury AKR, Chellappa R (2003) Face reconstruction from monocular video using uncertainty analysis and a generic model. Comput Vis Image Underst 91(1):188–213
Castrilln M, Dniz O, Guerra C, et al (2007) ENCARA2: Real-time detection of multiple faces at different resolutions in video streams. J Vis Commun Image Represent 18(2):130–140
Chouvatut V, Madarasmi S, Tuceryan M (2013) 3D face and motion estimation from sparse points using adaptive bracketed minimization. Multimedia tools and applications 63(2):569–589
Ding L, Ding X, Fang C (2014) 3D face sparse reconstruction based on local linear fitting. Vis Comput 30(2):189–200
Dai W, Milenkovic O (2010) SET: an algorithm for consistent matrix completion. IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP) 2010:3646–3649
Da F, Sui Y (2011) 3D reconstruction of human face based on an improved seeds-growing algorithm. Mach Vis Appl 22(5):879–887
Emrith K, Broadbent L, Smith LN, et al (2013) Real-time recovery of moving 3D faces for emerging applications. Comput Ind 64(9):1390–1398
Furukawa Y, Ponce J (2010) Accurate, dense, and robust multiview stereopsis. IEEE Trans Pattern Anal Mach Intell 32(8):1362–1376
Gonzalez-Mora J, De la Torre F, Guil N, et al (2010) Learning a generic 3D face model from 2D image databases using incremental structure-from-motion. Image Vis Comput 28(7):1117–1129
Hansen MF, Atkinson GA, Smith LN et al (2010) 3D face reconstructions from photometric stereo using near infrared and visible light. Comput Vis Image Underst 114(8):942–951
Herold C, Despiegel V, Gentric S, et al (2014) Recursive head reconstruction from multi-view video sequences. Comput Vis Image Underst 122:182–201
Jo J, Choi H, Kim IJ, et al (2015) Single-view-based 3D facial reconstruction method robust against pose variations. Pattern Recognit 48(1):73–85
Jones A, Lang M, Fyffe G, et al (2009) Achieving eye contact in a one-to-many 3D video teleconferencing system. ACM Trans Graph 28(3):64
Jain AK, Ross A, Prabhakar S (2004) An introduction to biometric recognition. IEEE Trans Circuits Syst Video Technol 14(1):4–20
Kemelmacher-Shlizerman I, Basri R (2011) 3d face reconstruction from a single image using a single reference face shape. IEEE Trans Pattern Anal Mach Intell 33(2):394–405
Kurtek S, Drira H (2015) A comprehensive statistical framework for elastic shape analysis of 3D faces. Comput Graph 51:52–59
Koo HS, Lam KM (2008) Recovering the 3d shape and poses of face images based on the similarity transform. Pattern Recogn Lett 29(6):712–723
Lin WY, Chen MY (2014) A novel framework for automatic 3D face recognition using quality assessment. Multimedia tools and applications 68(3):877–893
Lee SJ, Park KR, Kim J (2011) A SfM-based 3D face reconstruction method robust to self-occlusion by using a shape conversion matrix. Pattern Recogn 44(7):1470–1486
Lin Y, Medioni G, Choi J (2010) Accurate 3D face reconstruction from weakly calibrated wide baseline images with profile contours. IEEE Conf Comput Vis Pattern Recognit (CVPR) 2010:1490–1497
Lhuillier M, Quan L (2005) A quasi-dense approach to surface reconstruction from uncalibrated images. IEEE Trans Pattern Anal Mach Intell 27(3):418–433
Lin K, Wang X, Tan Y (2015) Self-adaptive morphable model based collaborative multi-view 3d face reconstruction in visual sensor network. Multimedia Tools and Applications 2015:1–23
Levine MD, Yu YC (2009) State-of-the-art of 3D facial reconstruction methods for face recognition based on a single 2D training image per person. Pattern Recogn Lett 30(10):908–913
Marques M, Costeira J (2009) Estimating 3D shape from degenerate sequences with missing data. Comput Vis Image Underst 113(2):261–272
Medioni G, Choi J, Kuo CH, et al (2009) Identifying noncooperative subjects at a distance using face images and inferred three-dimensional face models. IEEE Trans Syst Man Cybern Syst Hum 39(1):12–24
Maejima A, Wemler S, Machida T, et al (2008) Instant casting movie theater: the future cast system. IEICE Trans Inf Syst 91(4):1135–1148
Ohue K, Yamada Y, Uozumi S, et al (2006) Development of a new pre-crash safety system. SAE Technical Paper:2006
Peng W, Xu C, Feng Z (2016) 3D face modeling based on structure optimization and surface reconstruction with B-Spline. Neurocomputing 179:228–237
Rannar S, Lindgren F, Geladi P, et al (1994) A PLS kernel algorithm for data sets with many variables and fewer objects. Part 1: Theory and algorithm. J Chemom 8(2):111–125
Snchez-Escobedo D, Casteln M (2013) 3D face shape prediction from a frontal image using cylindrical coordinates and partial least squares. Pattern Recogn Lett 34(4):389–399
Snchez-Escobedo D, Casteln M, Smith WAP (2016) Statistical 3D face shape estimation from occluding contours. Comput Vis Image Underst 142:111–124
Sun Y, Dong J, Jian M, et al (2015) Fast 3D face reconstruction based on uncalibrated photometric stereo. Multimedia Tools and Applications 74(11):3635–3650
Schendel SA, Jacobson R, Khalessi S (2013) 3-dimensional facial simulation in orthognathic surgery: is it accurate?. J Oral Maxillofac Surg 71(8):1406–1414
Sun ZL, Lam KM, Gao QW (2013) Depth estimation of face images using the nonlinear least-squares model. IEEE Trans Image Process 22(1):17–30
Stylianou G, Lanitis A (2009) Image based 3D face reconstruction: a survey. International Journal of Image and Graphics 9(02):217–250
Song M, Tao D, Huang X, et al (2012) Three-dimensional face reconstruction from a single image by a coupled RBF network. IEEE Trans Image Process 21(5):2887–2897
Strecha C, von Hansen W, Gool LV, et al (2008) On benchmarking camera calibration and multi-view stereo for high resolution imagery. IEEE Conf Comput Vis Pattern Recognit (CVPR) 2008:1–8
Tomasi C, Kanade T (1992) Shape and motion from image streams under orthography: a factorization method. Int J Comput Vis 9(2):137–154
Tech M (2005) The BJUT-3D large-scale Chinese face database. Graphics Lab, Technical Report, Beijing University of Technology, 2005
Vandereycken B (2013) Low-rank matrix completion by Riemannian optimization. SIAM J Optim 23(2):1214–1236
Wang N, Gao X, Tao D, et al (2014) Facial feature point detection: A comprehensive survey. arXiv:1410.1037
Zeng D, Zhao Q, Long S, et al (2016) Examplar coherent 3D face reconstruction from forensic mugshot database. Image Vis Comput:2016
Acknowledgments
This paper is supported by National Natural Science Foundation of China under Grant #61472216, and by PhD Programs Foundation of Ministry of Education of China under Grant #20120002110067.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Dai, P., Wang, X. & Zhang, W. Coarse-to-fine multiview 3d face reconstruction using multiple geometrical features. Multimed Tools Appl 77, 939–966 (2018). https://doi.org/10.1007/s11042-016-4325-y
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-016-4325-y