Yong Zeng
Skip Abstract Section
Abstract
Facial recognition technology has been developed and widely used for decades. However, it has also made privacy concerns and researchers’ expectations for facial recognition privacy-preserving technologies. To provide privacy, detailed or semantic contents in face images should be obfuscated. However, face recognition algorithms have to be tailor-designed according to current obfuscation methods, as a result the face recognition service provider has to update its commercial off-the-shelf (COTS) products for each obfuscation method. Meanwhile, current obfuscation methods have no clearly quantified explanation. This paper presents a universal face obfuscation method for a family of face recognition algorithms using global or local structure of eigenvector space. By specific mathematical explanations, we show that the upper bound of the distance between the original and obfuscated face images is smaller than the given recognition threshold. Experiments show that the recognition degradation is 0% for global structure based and 0.3%-5.3% for local structure based, respectively. Meanwhile, we show that even if an attacker knows the whole obfuscation method, he/she has to enumerate all the possible roots of a polynomial with an obfuscation coefficient, which is computationally infeasible to reconstruct original faces. So our method shows a good performance in both privacy and recognition accuracy without modifying recognition algorithms.
- [1] . 2004. Face recognition technology: Security versus privacy. IEEE Technology and Society Magazine 23, 1 (2004), 9–19.Google Scholar
Cross Ref
- [2] . 2020. GraphInf: A GCN-based popularity prediction system for short video networks. In International Conference on Web Services. Springer, 61–76.Google ScholarDigital Library
- [3] . 2021. Face-off: Adversarial face obfuscation. Proc. Priv. Enhancing Technol. 2021, 2 (2021), 369–390.Google ScholarCross Ref
- [4] . 2020. Fawkes: Protecting privacy against unauthorized deep learning models. In 29th USENIX Security Symposium (USENIX Security 20). 1589–1604.Google Scholar
- [5] . 2019. Adversarial examples are not bugs, they are features. In Advances in Neural Information Processing Systems, Vol. 32.Google Scholar
- [6] . 2021. Identity-free facial expression recognition using conditional generative adversarial network. In 2021 IEEE International Conference on Image Processing (ICIP). IEEE, 1344–1348.Google ScholarCross Ref
- [7] . 2009. Large-scale privacy protection in Google Street View. In 2009 IEEE 12th International Conference on Computer Vision. IEEE, 2373–2380.Google ScholarCross Ref
- [8] . 2013. An ROI privacy protection scheme for H. 264 video based on FMO and chaos. IEEE Transactions on Information Forensics and Security 8, 10 (2013), 1688–1699.Google ScholarDigital Library
- [9] . 2007. PrivacyCam: A privacy preserving camera using uCLinux on the Blackfin DSP. In 2007 IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 1–8.Google ScholarCross Ref
- [10] . 2010. SCiFI-a system for secure face identification. In 2010 IEEE Symposium on Security and Privacy. IEEE, 239–254.Google ScholarDigital Library
- [11] . 1991. Eigenfaces for recognition. Journal of Cognitive Neuroscience 3, 1 (1991), 71–86.Google ScholarDigital Library
- [12] . 2014. Privacy-preserving outsourcing of image global feature detection. In 2014 IEEE Global Communications Conference. IEEE, 710–715.Google ScholarCross Ref
- [13] . 2009. Privacy-preserving face recognition. In International Symposium on Privacy Enhancing Technologies Symposium. Springer, 235–253.Google Scholar
- [14] . 2012. Image feature extraction in encrypted domain with privacy-preserving SIFT. IEEE Transactions on Image Processing 21, 11 (2012), 4593–4607.Google ScholarDigital Library
- [15] . 2016. A face template protection approach using chaos and GRP permutation. Security and Communication Networks 9, 18 (2016), 4957–4972.Google ScholarCross Ref
- [16] . 2006. Model-based face de-identification. In 2006 Conference on Computer Vision and Pattern Recognition Workshop (CVPRW’06). IEEE, 161–161.Google ScholarDigital Library
- [17] . 2013. Privacy-assured outsourcing of image reconstruction service in cloud. IEEE Transactions on Emerging Topics in Computing 1, 1 (2013), 166–177.Google ScholarCross Ref
- [18] . 2015. Privacy-preserving photo sharing based on a secure JPEG. In 2015 IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS). IEEE, 185–190.Google ScholarCross Ref
- [19] . 2016. iPrivacy: Image privacy protection by identifying sensitive objects via deep multi-task learning. IEEE Transactions on Information Forensics and Security 12, 5 (2016), 1005–1016.Google ScholarDigital Library
- [20] . 2013. Using face morphing to protect privacy. In 2013 10th IEEE International Conference on Advanced Video and Signal Based Surveillance. IEEE, 208–213.Google ScholarCross Ref
- [21] . 2013. Reversible data hiding in encrypted images by reserving room before encryption. IEEE Transactions on Information Forensics and Security 8, 3 (2013), 553–562.Google ScholarDigital Library
- [22] . 2014. Protected facial biometric templates based on local Gabor patterns and adaptive Bloom filters. In 2014 22nd International Conference on Pattern Recognition. IEEE, 4483–4488.Google ScholarDigital Library
- [23] . 2008. Face swapping: Automatically replacing faces in photographs. In ACM SIGGRAPH 2008 papers. 1–8.Google Scholar
- [24] . 2016. Defeating image obfuscation with deep learning. arXiv preprint arXiv:1609.00408 (2016).Google Scholar
- [25] . 2014. GARP-face: Balancing privacy protection and utility preservation in face de-identification. In IEEE International Joint Conference on Biometrics. IEEE, 1–8.Google ScholarCross Ref
- [26] . 2005. Preserving privacy by de-identifying face images. IEEE Transactions on Knowledge and Data Engineering 17, 2 (2005), 232–243.Google ScholarDigital Library
- [27] . 2018. Privacy protection in interactive content based image retrieval. IEEE Transactions on Dependable and Secure Computing 17, 3 (2018), 595–607.Google Scholar
- [28] . 2009. A hybrid approach for generating secure and discriminating face template. IEEE Transactions on Information Forensics and Security 5, 1 (2009), 103–117.Google ScholarDigital Library
- [29] . 2008. A new class of image registration for guaranteeing secure data management. In 2008 15th IEEE International Conference on Image Processing. IEEE, 269–272.Google ScholarCross Ref
- [30] . 2019. A privacy-preserving outsourcing scheme for image local binary pattern in secure industrial internet of things. IEEE Transactions on Industrial Informatics 16, 1 (2019), 629–638.Google ScholarCross Ref
- [31] . 2016. Face recognition in the scrambled domain via salience-aware ensembles of many kernels. IEEE Transactions on Information Forensics and Security 11, 8 (2016), 1807–1817.Google ScholarDigital Library
- [32] . 2020. A heterogeneous user authentication and key establishment for mobile client–server environment. Wireless Networks 26, 2 (2020), 913–924.Google ScholarDigital Library
- [33] . 2014. Deep learning face representation from predicting 10,000 classes. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 1891–1898.Google ScholarDigital Library
- [34] . 2015. FaceNet: A unified embedding for face recognition and clustering. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 815–823.Google ScholarCross Ref
- [35] . 2016. A discriminative feature learning approach for deep face recognition. In European Conference on Computer Vision. Springer, 499–515.Google ScholarCross Ref
- [36] . 2016. Large-margin softmax loss for convolutional neural networks. In ICML, Vol. 2. 7.Google Scholar
- [37] . 2018. CosFace: Large margin cosine loss for deep face recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 5265–5274.Google ScholarCross Ref
- [38] . 2019. ArcFace: Additive angular margin loss for deep face recognition. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 4690–4699.Google ScholarCross Ref
- [39] . 2002. Kernal Eigenfaces vs. Kernal Fisherfaces: Face recognition using kernal methods, automatrix face and gesture recognition, 202. In Proceedings, Fourth IEEE International Conference on. 208–213.Google Scholar
- [40] . 2000. Eigenhill vs. Eigenface and Eigenedge. In Proceedings 15th International Conference on Pattern Recognition. ICPR-2000, Vol. 2. IEEE, 827–830.Google ScholarCross Ref
- [41] . 2004. Two-dimensional PCA: A new approach to appearance-based face representation and recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence 26, 1 (2004), 131–137.Google ScholarDigital Library
- [42] . 2001. PCA versus LDA. IEEE Transactions on Pattern Analysis and Machine Intelligence 23, 2 (2001), 228–233.Google ScholarDigital Library
- [43] . 2005. Face recognition using Laplacianfaces. IEEE Transactions on Pattern Analysis and Machine Intelligence 27, 3 (2005), 328–340.Google ScholarDigital Library
- [44] . 2006. Face recognition using discriminant locality preserving projections. Image and Vision Computing 24, 3 (2006), 239–248.Google ScholarDigital Library
- [45] . 2008. Fast iterative solution of integral equations with method of moments and matrix decomposition algorithm–singular value decomposition. IEEE Transactions on Antennas and Propagation 56, 8 (2008), 2314–2324.Google ScholarCross Ref
- [46] . 2017. Key independent encrypted face clustering. In 2017 IEEE International Conference on Industrial and Information Systems (ICIIS). IEEE, 1–6.Google ScholarCross Ref
- [47] . 2011. Binary discriminant analysis for generating binary face template. IEEE Transactions on Information Forensics and Security 7, 2 (2011), 613–624.Google ScholarDigital Library
- [48] . 2010. Cancelable templates for sequence-based biometrics with application to on-line signature recognition. IEEE Transactions on Systems, Man, and Cybernetics-Part A: Systems and Humans 40, 3 (2010), 525–538.Google ScholarDigital Library
- [49] . 2019. Efficient obfuscation for CNF circuits and applications in cloud computing. Soft Computing 23, 6 (2019), 2061–2072.Google ScholarDigital Library
- [50] . 2003. SIFT: Predicting amino acid changes that affect protein function. Nucleic Acids Research 31, 13 (2003), 3812–3814.Google ScholarCross Ref
- [51] . 2014. Towards efficient privacy-preserving image feature extraction in cloud computing. In Proceedings of the 22nd ACM International Conference on Multimedia. 497–506.Google ScholarDigital Library
- [52] . 2012. Image feature extraction in encrypted domain with privacy-preserving SIFT. IEEE Transactions on Image Processing 21, 11 (2012), 4593–4607.Google ScholarDigital Library
- [53] . 2016. Securing SIFT: Privacy-preserving outsourcing computation of feature extractions over encrypted image data. IEEE Transactions on Image Processing 25, 7 (2016), 3411–3425.Google ScholarDigital Library
- [54] . 2020. Deep multi-label learning for image distortion identification. Signal Processing 172 (2020), 107536.Google ScholarCross Ref
- [55] . 2022. A Survey on Image Deblurring. (2022).
arxiv:cs.CV/2202.07456 Google Scholar - [56] . 2022. DuetFace: Collaborative privacy-preserving face recognition via channel splitting in the frequency domain. In Proceedings of the 30th ACM International Conference on Multimedia. 6755–6764.Google ScholarDigital Library
- [57] . 2022. Privacy-preserving face recognition in the frequency domain. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 36. 2558–2566.Google ScholarCross Ref
- [58] . 2022. Privacy protection framework for face recognition in edge-based Internet of Things. Cluster Computing (2022), 1–19.Google Scholar
- [59] . 2023. PriFace: A privacy-preserving face recognition framework under untrusted server. Journal of Ambient Intelligence and Humanized Computing (2023), 1–13.Google Scholar
- [60] . 2020. Privacy preserving face recognition utilizing differential privacy. Computers & Security 97 (2020), 101951.Google ScholarDigital Library
- [61] . 2022. A differential privacy-based classification system for edge computing in IoT. Computer Communications 182 (2022), 117–128.Google ScholarDigital Library
- [62] . 2022. Privacy-preserving face recognition with learnable privacy budgets in frequency domain. In Computer Vision–ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XII. Springer, 475–491.Google ScholarDigital Library
- [63] . 2016. Face recognition in the scrambled domain via salience-aware ensembles of many kernels. IEEE Transactions on Information Forensics and Security 11, 8 (2016), 1807–1817.Google ScholarDigital Library
- [64] . 1997. Eigenfaces vs. Fisherfaces: Recognition using class specific linear projection. IEEE Transactions on Pattern Analysis and Machine Intelligence 19, 7 (1997), 711–720.Google ScholarDigital Library
- [65] . 2015. When face recognition meets with deep learning: An evaluation of convolutional neural networks for face recognition. In Proceedings of the IEEE International Conference on Computer Vision Workshops. 142–150.Google ScholarDigital Library
- [66] . 1970. Positive definite matrices. The American Mathematical Monthly 77, 3 (1970), 259–264.Google ScholarCross Ref
- [67] . 1999. Learning the parts of objects by non-negative matrix factorization. Nature 401, 6755 (1999), 788–791.Google ScholarCross Ref
- [68] . 2010. Image quality assessment: A multiscale geometric analysis-based framework and examples. (2010).Google Scholar
- [69] . 2019. FaceQNet: Quality assessment for face recognition based on deep learning. In 2019 International Conference on Biometrics (ICB). IEEE, 1–8.Google ScholarCross Ref
- [70] . 2022. Performance face image quality assessment under the difference of illumination directions in face recognition system using FaceQNet, SDD-FIQA, and SER-FIQ. In The 2022 International Conference on Computer, Control, Informatics and Its Applications. 219–223.Google ScholarDigital Library
- [71] . 2021. The effect of wearing a face mask on face image quality. In 2021 16th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2021). IEEE, 1–8.Google ScholarDigital Library
- [72] . 2020. An efficient method for face quality assessment on the edge. In Computer Vision–ECCV 2020 Workshops: Glasgow, UK, August 23–28, 2020, Proceedings, Part V 16. Springer, 54–70.Google ScholarDigital Library
- [73] . 1991. Recursive updating the eigenvalue decomposition of a covariance matrix. IEEE Transactions on Signal Processing 39, 5 (1991), 1136–1145.Google ScholarDigital Library
- [74] . 2017. The complexity of differential privacy. Tutorials on the Foundations of Cryptography: Dedicated to Oded Goldreich (2017), 347–450.Google ScholarCross Ref
- [75] . 2003. Computational complexity. In Encyclopedia of Computer Science. 260–265.Google Scholar
- [76] . 2017. Linear Algebra with Applications. Jones & Bartlett Learning.Google Scholar
Index Terms
- Eyes See Hazy while Algorithms Recognize Who You Are
Recommendations
Towards efficient privacy-preserving face recognition in the cloud
Highlights- We propose a randomness-based privacy-preserving face recognition scheme.
- We ...
AbstractFace recognition (FR) has become increasingly significant in many computer vision applications. However, with the rapid deployment of FR, the privacy of face images has been a growing concern, especially when FR is performed in cloud ...
3D Face Factorisation for Face Recognition Using Pattern Recognition Algorithms
AbstractThe face is the preferable biometrics for person recognition or identification applications because person identifying by face is a human connate habit. In contrast to 2D face recognition, 3D face recognition is practically robust to illumination ...
PRO-Face: A Generic Framework for Privacy-preserving Recognizable Obfuscation of Face Images
MM '22: Proceedings of the 30th ACM International Conference on MultimediaA number of applications (e.g., video surveillance and authentication) rely on automated face recognition to guarantee functioning of secure services, and meanwhile, have to take into account the privacy of individuals exposed under camera systems. This ...
Comments