Adaptive Privacy-Preserving Techniques for Multimedia Content Processing in Cloud Environments: A Differential Privacy Approach
Main Article Content
Abstract
We propose a novel adaptive differential privacy framework for multimedia content processing in cloud environments, designed to achieve optimal privacy-utility trade-offs through content-aware noise calibration and dynamic budget allocation. The framework introduces three core technical innovations: (1) a sensitivity-guided privacy budget allocation mechanism that reduces utility loss by 38.7% compared to uniform allocation, (2) a frequency-domain noise injection strategy that preserves perceptual quality while ensuring epsilon-differential privacy, and (3) an optimization algorithm that solves the budget allocation problem in O (n log n) time. Extensive experiments on the COCO, AudioSet, and UCF101 datasets demonstrate that the proposed framework maintains 91.3% task accuracy at epsilon = 1.0 while reducing membership inference attack success rates to 52.8%. Moreover, the system processes up to 312 images per second on commodity hardware, underscoring its practicality for deployment in large-scale production cloud environments.
Article Details
Section
How to Cite
References
1. Y. Zhao, and J. Chen, "A survey on differential privacy for unstructured data content," ACM Computing Surveys (CSUR), vol. 54, no. 10s, pp. 1-28, 2022. doi: 10.1145/3490237
2. J. So, B. Güler, and A. S. Avestimehr, "CodedPrivateML: A fast and privacy-preserving framework for distributed machine learning," IEEE Journal on Selected Areas in Information Theory, vol. 2, no. 1, pp. 441-451, 2021.
3. A. K. Singh, and R. Gupta, "A privacy-preserving model based on differential approach for sensitive data in cloud environment," Multimedia Tools and Applications, vol. 81, no. 23, pp. 33127-33150, 2022. doi: 10.1007/s11042-021-11751-w
4. Z. Zhou, H. Chen, L. Chen, D. Zhang, C. Wu, X. Liu, and M. K. Khan, "NetDP: In-network differential privacy for large-scale data processing," IEEE Transactions on Green Communications and Networking, 2024. doi: 10.1109/tgcn.2024.3432781
5. A. Rahdari, E. Keshavarz, E. Nowroozi, R. Taheri, M. Hajizadeh, M. Mohammadi, and T. Bauschert, "A survey on privacy and security in distributed cloud computing: Exploring federated learning and beyond," IEEE Open Journal of the Communications Society, 2025. doi: 10.1109/ojcoms.2025.3560034
6. F. Amiri, G. Quirchmayr, and E. Weippl, "Introducing differential privacy to recommender systems through a self-organizing feature map using a local distance-based methodology," IEEE Access, 2025. doi: 10.1109/access.2025.3605563
7. S. Y. Chang, H. C. Wu, K. Yan, S. C. H. Huang, and Y. Wu, "Federated multimedia recommendation systems with privacy protection," In 2024 IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB), June, 2024, pp. 1-7.
8. G. Zhu, X. Li, C. Zheng, and L. Wang, "Multimedia fusion privacy protection algorithm based on IoT data security under network regulations," Computational Intelligence and Neuroscience, vol. 2022, no. 1, p. 3574812, 2022.
9. X. Wang, J. Lv, B. G. Kim, C. Maple, B. D. Parameshachari, A. Slowik, and K. Li, "Generative adversarial privacy for multimedia analytics across the IoT-edge continuum," IEEE Transactions on Cloud Computing, 2024.
10. B. Liu, M. Ding, S. Shaham, W. Rahayu, F. Farokhi, and Z. Lin, "When machine learning meets privacy: A survey and outlook," ACM Computing Surveys (CSUR), vol. 54, no. 2, pp. 1-36, 2021.
11. R. Xu, N. Baracaldo, and J. Joshi, "Privacy-preserving machine learning: Methods, challenges and directions," arXiv preprint arXiv:2108.04417, 2021.
12. A. R. Shahid, and A. Imteaj, "Securing user privacy in cloud-based whiteboard services against health attribute inference attacks," IEEE Transactions on Artificial Intelligence, vol. 5, no. 8, pp. 3872-3885, 2024. doi: 10.1109/tai.2024.3352529
13. R. Gupta, and A. K. Singh, "A differential privacy-based secure data sharing model in cloud environment," In 2022 IEEE 6th Conference on Information and Communication Technology (CICT), November, 2022, pp. 1-6. doi: 10.1109/cict56698.2022.9997953
14. M. A. Husnoo, A. Anwar, R. K. Chakrabortty, R. Doss, and M. J. Ryan, "Differential privacy for IoT-enabled critical infrastructure: A comprehensive survey," IEEE Access, vol. 9, pp. 153276-153304, 2021. doi: 10.1109/access.2021.3124309
15. Z. Wang, Z. Liu, Y. Luo, T. Zhou, J. Qin, and Z. Cai, "PPIDM: Privacy-preserving inference for diffusion model in the cloud," IEEE Transactions on Circuits and Systems for Video Technology, 2025. doi: 10.1109/tcsvt.2025.3553514