DR-DPGAN: Dual Prior GAN for Image Privacy Based on Dynamic Reversibility

Tingting Lin; Tao Wang; Zhigao Zheng

doi:10.53941/jadc.2025.100001

Abstract

With the development of image acquisition technology, the volume of image data has surged, highlighting the contradiction between data publication and privacy protection. Generative Adversarial Networks (GANs) offer a solution to find a balance between the development of image data and privacy security. However, the unidirectional image generation of GANs fails to satisfy the reversible requirements of privacy-sensitive images. To address this limitation, this study proposes an image privacy protection method based on a dual-prior GAN with dynamic reversibility, called DR-DPGAN. This method uses StyleGAN2 latent space editing to make targeted modifications to image identity features, allowing modification and reconstruction of features. To achieve image privacy protection, a fake identity generator composed of two-layer multi-layer perceptrons is designed. By combining identity-related guidance information, it precisely controls the generation of fake features to avoid excessive or insufficient modification. Meanwhile, three-dimensional prior constraints are introduced to extract geometric feature vectors, maximizing the retention of original non-identity attribute features and ensuring the usability of images in downstream tasks. To ensure reversible image restoration, this paper converts the original identity attribute information into binary vectors through a binary encoding mapping network, generating reversible encrypted features to ensure precise restoration of original identity features. In addition, four loss functions are used jointly to optimize the network to balance the quality of the generation. To verify the reversibility and effectiveness of the proposed method, comprehensive experimental tests are conducted on two different datasets. The experimental results demonstrate the effectiveness of this method in image anonymization and reversible restoration.

References

1.
Morris, J.; Newman, S.; Palaniappan, K.; et al. Do you know you are tracked by photos that you didn’t take: Large-scale location-aware multi-party image privacy protection. IEEE Trans. Dependable Secur. Comput. 2023, 20, 301–312.
2.
Chunling, H.; Rui, X. Differentially private gans by adding noise to discriminator’s loss. Comput. Secur. 2021, 107, 102322–102332.
3.
Jindong, J.; Wafa, S.; Ali, S.; Laurent, G. Effect of face blurring on human pose estimation: Ensuring subject privacy for medical and occupational health applications. Sensors 2022, 22, 9376.
4.
Kumar, A.R.; Sharma, A.K. Reversible data hiding in encrypted image using bit-plane based label-map encoding with optimal block size. J. Inf. Secur. Appl. 2025, 90, 104005.
5.
Kumar, A.R.; Sharma, A.K.; Ranjan, P. High-capacity reversible data hiding in encrypted images based on difference image transfiguration. Signal Image Video Process. 2025, 19, 365.
6.
Konduru, U.R.; Nagarajan, A.P.; Sai, C.V.S. An improved performance of reversible data hiding in encrypted images using decision tree algorithm. Eng. Appl. Artif. Intell. 2024, 137, 109100.
7.
Kim, B.N.; Dolz, J.; Jodoin, P.M.; et al. Privacy-net: An adversarial approach for identity-obfuscated segmentation of medical images. IEEE Trans. Med. Imaging 2021, 40, 1737–1749.
8.
Liao, X.; Wang, Y.; Wang, T.; et al. Famm: Facial muscle motions for detecting compressed deepfake videos over social networks. IEEE Trans. Circuits Syst. Video Technol. 2023, 33, 7236–7251.
9.
Fu, L.; Liao, X.; Guo, J.; Dong, L.; Qin, Z. Waverecovery: Screen-shooting watermarking based on wavelet and recovery. IEEE Trans. Circuits Syst. Video Technol. 2025, 35, 3603–3618.
10.
Li, Y.; Liao, X.; Wu, X. Screen-shooting resistant watermarking with grayscale deviation simulation. IEEE Trans. Multimed. 2024, 26, 10908–10923.
11.
Yang, Y.; He, H.; Chen, F.; et al. Secure reversible data hiding in encrypted image based on 2d labeling and block classification coding. J. Inf. Secur. Appl. 2024, 83, 103771–103780.
12.
Liao, X.; Yin, J.; Chen, M.; Qin, Z. Adaptive payload distribution in multiple images steganography based on image texture features. IEEE Trans. Dependable Secur. Comput. 2022, 19, 897–911.
13.
Wu, H.; Cheung, Y.; Zhuang, Z.; Xu, L.; Hu, J. Lossless data hiding in encrypted images compatible with homomorphic processing. IEEE Trans. Cybern. 2022, 53, 3688–3701.
14.
Xiong, J.; Chen, J.; Lin, J.; Jiao, D.; Liu, H. Enhancing privacy-preserving machine learning with self-learnable activation functions in fully homomorphic encryption. J. Inf. Secur. Appl. 2024, 86, 103887.
15.
Bai, Y.; Zhao, H.; Shi, X.; Chen, L. Towards practical and privacy-preserving cnn inference service for cloud-based medical imaging analysis: A homomorphic encryption-based approach. Comput. Methods Programs Biomed. 2025, 261, 108599.
16.
Gu, X.; Luo, W.; Ryoo, M.S.; et al. Password-conditioned anonymization and deanonymization with face identity transform- ers. In Proceedings of the of European Conference on Computer Vision, Glasgow, UK, 23–28 August 2020; pp. 727–743.
17.
Zhang, Y.; Wang, T.; Zhao, R.; et al. Rapp: Reversible privacy preservation for various face attributes. IEEE Trans. Inf. Forensics Secur. 2023, 18, 3074–3087.
18.
Su, Z.; Zhang, G.; Shi, Z.; et al. Message-driven generative music steganography using midi-gan. IEEE Trans. Dependable Secur. Comput. 2024, 21, 5196–5207.
19.
Song, J.; Ye, J.-C. Federated cyclegan for privacy-preserving image-to-image translation. arXiv 2021, arXiv:2106.09246.
20.
Li, D.; Wang, W.; Zhao, K.; et al. Riddle: Reversible and diversified de-identification with latent encryptor. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, BC, Canada, 18–22 June 2023; pp. 8093–8102.
21.
Karras, T.; Laine, S.; Aittala, M.; et al. Analyzing and improving the image quality of stylegan. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA, 14–19 June 2020; pp. 8107–8116.

Scilight Press

Author Information

Abstract

Keywords

References

About Scilight

Journals

Publishing Policies

Contact Us