从图像视频重建人的3D形状与识别技术研究

信息安全研究 ›› 2019, Vol. 5 ›› Issue (10): 935-943.

从图像视频重建人的3D形状与识别技术研究

张立武¹,王蕊²,高红丽³,田启川⁴,何荣祖⁴,马凌慧⁴,王蕊³

1. 中国科学院软件研究所可信计算与信息保障实验室
2. 信息安全国家重点实验室(中国科学院信息工程研究所)
3. 北京工业大学应用数理学院
4. 北京建筑大学电气与信息工程学院

收稿日期:2019-10-08 出版日期:2019-10-15 发布日期:2019-10-08
通讯作者: 张立武
作者简介:张立武，男，1976年生，博士学位，中国科学院软件研究所正高级工程师/博士生导师，主要研究领域为网络空间安全、计算机视觉、人工智能，E-mail: liwu@iscas.ac.cn 王蕊，女，1981年生，博士学位，中国科学院信息工程研究所研究员/博士生导师，主要研究领域为计算机视觉、人工智能，E-mail: wangrui@iie.ac.cn. 高红丽，女，1986年生，博士学位，北京工业大学应用数理学院讲师，主要研究领域为应用物理、光学，E-mail: hlgao@bjut.edu.cn. 田启川，男，1971年生，博士学位，北京建筑大学电气与信息工程学院教授，主要研究领域为人工智能、生物特征识别、计算机视觉与图像处理，E-mail: tianqichuan@bucea.edu.cn. 何荣祖，女，1997年生，北京建筑大学电气与信息工程学院在读本科生，E-mail: aria1023@163.com. 马凌慧，女，1996年生，北京建筑大学电气与信息工程学院在读本科生，E-mail: linghui1126@163.com. 王蕊，女，1997年生，北京工业大学应用数理学院在读本科生，E-mail: Wangruiluckygo@163.com.

Research on Reconstruction and Recognition of Human 3D Shapes from Images and Videos

Received:2019-10-08 Online:2019-10-15 Published:2019-10-08

摘要/Abstract

摘要： 基于生物特征的身份识别前景光明，然而还存在诸多技术问题阻碍其应用：如比较流行的人脸识别技术还没有达到理想的精度，单一的生物特征也容易受遮挡、姿态等各种因素影响.提出了一种从2D图像视频重建3D形状并加以识别的技术思路，利用大量植入的3D形状知识约束，结合分类决策算法分析，评估2D图像视频，完成图像目标定位、3D形状重建与识别.此外，还对如何解决远程人脸识别中的“换脸攻击”进行了探讨.

关键词: 人脸识别, 3D重建, 2D图像, 知识约束, 面向对象

Abstract: Biometric-based Authentication has a bright future. However, there are still many technical problems to be solved in its application: for example, face recognition technology has not achieved the desired accuracy, and recognition based on a single biometric is susceptible to various factors such as occlusion and pose. We proposed a technical idea of reconstructing 3D shapes from 2D imagesvideos and recognition. Using a large number of implanted 3D shape knowledge constraints, combined with classification decision algorithm to analyze and evaluate 2D imagesvideos to finish object detection and alignment, 3D shape reconstruction and recognition. In addition, we also discussed how to solve the “face-swap attack” in remote face recognition.

Key words: face recognition, 3D reconstruction, 2D images, knowledge constraints, object oriented

张立武王蕊高红丽田启川何荣祖马凌慧王蕊. 从图像视频重建人的3D形状与识别技术研究[J]. 信息安全研究, 2019, 5(10): 935-943.

参考文献

[1] 方淑仙, 张立武, 王惠莅. 基于人脸生物特征的身份鉴别技术发展趋势研究[J]. 信息安全研究, 2017(6). [2] Su J, Vargas D V, Kouichi S. One pixel attack for fooling deep neural networks[J]. arXiv preprint arXiv:1710.08864, 2017. [3] Turk M, Pentland A. Eigenfaces for recognition[J]. Journal of cognitive neuroscience, 1991, 3(1): 71-86. [4] Belhumeur P N, Hespanha J P, Kriegman D J. Eigenfaces vs. fisherfaces: Recognition using class specific linear projection[J]. IEEE Transactions on pattern analysis and machine intelligence, 1997, 19(7): 711-720. [5] He X, Yan S, Hu Y, et al. Face recognition using laplacianfaces[J]. IEEE transactions on pattern analysis and machine intelligence, 2005, 27(3): 328-340. [6] Cho H, Roberts R, Jung B, et al. An efficient hybrid face recognition algorithm using pca and gabor wavelets[J]. International Journal of Advanced Robotic Systems, 2014, 11(4): 59. [7] Lades M, Vorbruggen J C, Buhmann J, et al. Distortion invariant object recognition in the dynamic link architecture[J]. IEEE Transactions on computers, 1993, 42(3): 300-311. [8] Osuna E, Freund R, Girosit F. Training support vector machines: an application to face detection[C]//Computer vision and pattern recognition, 1997. Proceedings., 1997 IEEE computer society conference on. IEEE, 1997: 130-136. [9] Huang G B, Lee H, Learned-Miller E. Learning hierarchical representations for face verification with convolutional deep belief networks[C]//Computer Vision and Pattern Recognition (CVPR), 2012 IEEE Conference on. IEEE, 2012: 2518-2525. [10] Lu C, Tang X. Surpassing Human-Level Face Verification Performance on LFW with GaussianFace[J]. Computer Science, 2014. [11] Taigman Y, Yang M, Ranzato M A, et al. Deepface: Closing the gap to human-level performance in face verification[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2014: 1701-1708. [12] Ouyang W, Zeng X, Wang X, et al. DeepID-Net: Deformable deep convolutional neural networks for object detection[J]. IEEE Trans on Pattern Analysis & Machine Intelligence, 2014, 46(5):2403-2412. [13] Schroff F, Kalenichenko D, Philbin J. Facenet: A unified embedding for face recognition and clustering[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2015: 815-823. [14] Parke F I. Computer generated animation of faces[C]//Proceedings of the ACM annual conference-Volume 1. ACM, 1972: 451-457. [15] Blanz V, Vetter T. A morphable model for the synthesis of 3D faces[C]//Proceedings of the 26th annual conference on Computer graphics and interactive techniques. ACM Press/Addison-Wesley Publishing Co., 1999: 187-194. [16] Paysan P, Knothe R, Amberg B, et al. A 3D face model for pose and illumination invariant face recognition[C]//Advanced video and signal based surveillance, 2009. AVSS'09. Sixth IEEE International Conference on. IEEE, 2009: 296-301. [17] Kaashki N N, Safabakhsh R. 3D constrained local model-based face recognition using Kinect under variant conditions[C]// International Symposium on Telecommunications. IEEE, 2014:361-366. [18] Ikeuchi K, Horn B K P. Numerical shape from shading and occluding boundaries[J]. Artificial intelligence, 1981, 17(1-3): 141-184. [19] Kemelmacher-Shlizerman I, Basri R. 3D face reconstruction from a single image using a single reference face shape[J]. IEEE transactions on pattern analysis and machine intelligence, 2011, 33(2): 394-405. [20] Fidaleo D, Medioni G. Model-assisted 3d face reconstruction from video[C]//International Workshop on Analysis and Modeling of Faces and Gestures. Springer Berlin Heidelberg, 2007: 124-138. [21] Zhang L, Snavely N, Curless B, et al. Spacetime faces: high resolution capture for modeling and animation[C]//ACM Transactions on Graphics (TOG). ACM, 2004, 23(3): 548-558. [22] Siebert J P, Urquhart C W. C3D™: a Novel Vision-Based 3-D Data Acquisition System[M]//Image Processing for Broadcast and Video Production. Springer London, 1995: 170-180. [23] Beeler T, Bickel B, Beardsley P, et al. High-quality single-shot capture of facial geometry[C]//ACM Transactions on Graphics (ToG). ACM, 2010, 29(4): 40. [24] Zhang P B, Hung Y S. Non-Rigid Structure from Motion through Estimation of Blend Shapes[C]//Digital Image Computing: Techniques and Applications (DICTA), 2015 International Conference on. IEEE, 2015: 1-7. [25] Zhu X, Lei Z, Yan J, et al. High-fidelity pose and expression normalization for face recognition in the wild[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015: 787-796. [26] S. Platt,N. Badler, Animating Facial Expressions, Proc. SIGGRAPH'81, 1981, vol.15: 245-252. [27] I. Essa, S. Basu, Modeling, Tracking and Interactive Animation of Facial Expressions and Head Movements Using Input from Video, "Proc. Computer Animation. June 1996. 68-79. [28] D. Terzopoulus, K. Waters, Physically-Base Facial Modeling, Analysis and Animation, [J]. Visualization and Computer Animation, 1990, vol 1: 73-80. [29] M. Bro-Nielsen, S. Cotin, Real-Time Volumetric Deformable Models for Surgery Simulation Using Finite Elements and Condensatio, Proc. EUROGRAPHICS'96, Aug 1996,vol 15:57-66. [30] S. Cotin, H. Delingette, N. Ayache, Real-Time Elastic Deformations of Soft Tissues for Surgery Simulation, IEEE Trans. Visualization and Computer Graphics, Jan -Mar 1999, vol 5,no.1:62-73. [31] Smolyanskiy N, Huitema C, Liang L, et al. Real-time 3D face tracking based on active appearance model constrained by depth data [J]. Image & Vision Computing, 2014, 32(11):860-869. [32] Thies J, Stamminger M, Theobalt C. Demo of Face2Face: real-time face capture and reenactment of RGB videos[C]// IEEE Conference on Computer Vision and Pattern Recognition. United States of America：IEEE Computer society, 2016:2387-2395. [33] Liang D , Wang R , Tian X , et al. PCGAN: Partition-Controlled Human Image Generation[J]. 2018. [34] Albert Pumarola, Antonio Agudo, Aleix M.Martinez, Alberto Sanfeliu, GANimation: Anatomically-aware Facial Animation from a Single Image, ECCV,2018. [35] 张立武, 王惠莅. 设备与服务的可信连接安全要求国际标准提案研究[J]. 信息技术与标准化, 2018, No.401(05):48-51+57.