ACM Multimedia 2020 Tutorial

Deep Learning for Privacy in Multimedia

Andrea Cavallaro, Mohammad Malekzadeh, Ali Shahin Shamsabadi


Privacy Multimedia

We discuss the design and evaluation of machine learning algorithms that provide users with more control on the multimedia information they share. We introduce privacy threats for multimedia data and key features of privacy protection. We cover privacy threats and mitigating actions for images, videos, and motion-sensor data from mobile and wearable devices, and their protection from unwanted, automatic inferences.

Slides: Part1, Part2, Part3.


Resources

Code and data discussed and demoed during the tutorial:


References

[1] M.S. Cross, A. Cavallaro. 2020. Privacy as a feature for body worn cameras. Signal Processing Magazine, 37 (4), doi: 10.1109/MSP.2020.2989686.

[2] A.S. Shamsabadi, R.S. Matilla, A. Cavallaro. 2020. ColorFool: Semantic Adversarial Colorization. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, Washington, USA.

[3] G. Abebe Tadesse, O. Bent, L. Marcenaro, K. Weldemariam, A. Cavallaro. 2020. Privacy-aware human activity recognition from a wearable camera. Signal Processing Magazine, 37 (3), doi: 10.1109/MSP.2020.2976158.

[4] A.S. Shamsabadi, C. Oh, A. Cavallaro. 2020. EdgeFool: An Adversarial Image Enhancement Filter. In Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain.

[5] R. S. Matilla, C. Y. Li, A. S. Shamsabadi, R. Mazzon, A. Cavallaro. 2020. Exploiting vulnerabilities of deep neural networks for privacy protection. IEEE Transactions on Multimedia, 22 (7), doi: 10.1109/TMM.2020.2987694.

[6] A. S. Shamsabadi, A. Gascon, H. Haddadi, A. Cavallaro. 2020. PrivEdge: From Local to Distributed Private Training and Prediction. IEEE Transactions on Information Forensics and Security, 15, doi: 10.1109/TIFS.2020.2988132.

[7] S. A. Osia, A. S. Shamsabadi, S. Sajadmanesh, A. Taheri, K. Katevas, H. R. Rabiee, N. D. Lane, H. Haddadi. 2020. A Hybrid Deep Learning Architecture for Privacy-Preserving Mobile Analytics. IEEE Internet of Things Journal.

[8] M. Malekzadeh, R. G. Clegg, A. Cavallaro, H. Haddadi. 2020. Privacy and Utility Preserving Sensor Data Transformations. Pervasive and Mobile Computing, Volume 63, Article 101132.

[9] M. Malekzadeh, D. Athanasakis, H. Haddadi, B. Livshits. 2020. Privacy-Preserving Bandits. In Proceedings of the Conference on Machine Learning and Systems (MLSys), Huston, Texas, USA.

[10] O. Sarwar, B. Rinner, A. Cavallaro. 2019. A privacy-preserving filter for oblique face images based on adaptive hopping Gaussian mixtures. IEEE Access, 7, doi: 10.1109/ACCESS.2019.2944861.

[11] C. Y. Li, A. S. Shamsabadi, R. S. Matilla, R. Mazzon, A. Cavallaro. 2019. Scene Privacy Protection. In Proc. of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, UK.

[12] M. Malekzadeh, R. G. Clegg, A. Cavallaro, H. Haddadi. 2019. Mobile Sensor Data Anonymization. In Proc. of the IEEE/ACM International Conference on Internet of Things Design and Implementation (IoTDI), Montreal, Canada.

[13] A. S. Shamsabadi, H. Haddadi, A. Cavallaro. 2018. Distributed One-Class Learning. In Proceedings of the IEEE International Conference on Image Processing (ICIP), Athens, Greece.

[14] S. A. Osia, A. Taheri, A. S. Shamsabadi, K. Katevas, H. Haddadi, H. R. Rabiee. 2018. Deep Private-Feature Extraction. IEEE Transactions on Knowledge and Data Engineering.

[15] S. A. Osia, A. S. Shamsabadi, A. Taheri, H. R. Rabiee, H. Haddadi. 2018. Private and Scalable Personal Data Analytics using Hybrid Edge-Cloud Deep Learning. IEEE Computer, Special Issue on Mobile and Embedded Deep Learning.

[16] M. Malekzadeh, R. G. Clegg, A. Cavallaro, H. Haddadi. 2018. Protecting Sensory Data Against Sensitive Inferences. In Proc. of the 1st ACM Workshop on Privacy by Design in Distributed Systems (W-P2DS), Porto, Portugal.

[17] M. Malekzadeh, R. G. Clegg, H. Haddadi. 2018. Replacement Autoencoder: A Privacy-Preserving Algorithm for Sensory Data Analysis. In Proceedings of the IEEE/ACM International Conference on Internet-of-Things Design and Implementation (IoTDI), Orlando, Florida, USA.

[18] O. Sarwar, A. Cavallaro, B. Rinner. 2018. Temporally smooth privacy protected airborne videos. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Madrid, Spain.


Contacts

Andrea Cavallaro, a.cavallaro@qmul.ac.uk

Mohammad Malekzadeh, m.malekzadeh@qmul.ac.uk

Ali Shahin Shamsabadi, a.shahinshamsabadi@qmul.ac.uk


Acknowledgments

We wish to thank the Alan Turing Institute (EP/N510129/1), which is funded by the U.K. Engineering and Physical Sciences Research Council, for its support throughout the project Privacy-Preserving Multimodal Learning for Activity Recognition (PRIMULA).