SetRkNN: Efficient and Privacy-Preserving Set Reverse kNN Query in Cloud

Zheng, Yandong; Lu, Rongxing; Zhu, Hui; Zhang, Songnian; Guan, Yunguo; Shao, Jun; Wang, Fengwei; Li, Hui

doi:10.1109/tifs.2022.3231785

Cited by 12 publications

(1 citation statement)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Yuan et al [24] devises an Identifiable Virtual Face Generator (IVFG) to produce the virtual faces that are recognizable via their virtual ID embeddings. Wen et al [25] attempts to protect facial privacy in video frames via a modular architecture named IdentityMask, which leverages deep motion flow and protective motion IdentityMask, Zhang et al [26] proposes RAPP, a reversible privacy-preserving scheme for protecting various facial attributes. Recently, Li et al [27] proposes RiDDLE, a reversible DeID framework with StyleGAN [28] latent space encryption.…”

Section: B Deep-based Facial Anonymizationmentioning

confidence: 99%

PRO-Face: A Generic Framework for Privacy-preserving Recognizable Obfuscation of Face Images

Lin

Liu

et al. 2022

Proceedings of the 30th ACM International Conference on Multimedia

View full text Add to dashboard Cite

This paper proposes a novel paradigm for facial privacy protection that unifies multiple characteristics including anonymity, diversity, reversibility and security within a single lightweight framework. We name it PRO-Face S, short for Privacy-preserving Reversible Obfuscation of Face images via Secure flow-based model. In the framework, an Invertible Neural Network (INN) is utilized to process the input image along with its pre-obfuscated form, and generate the privacy protected image that visually approximates to the pre-obfuscated one, thus ensuring privacy. The pre-obfuscation applied can be in diversified form with different strengths and styles specified by users. Along protection, a secret key is injected into the network such that the original image can only be recovered from the protection image via the same model given the correct key provided. Two modes of image recovery are devised to deal with malicious recovery attempts in different scenarios. Finally, extensive experiments conducted on three public image datasets demonstrate the superiority of the proposed framework over multiple state-of-the-art approaches.

show abstract