Privacy-Preserving Video Classification with Convolutional Neural Networks

Sikha Pentyala, Rafael Dowsley, Martine De Cock
Proceedings of the 38th International Conference on Machine Learning, PMLR 139:8487-8499, 2021.

Abstract

Many video classification applications require access to personal data, thereby posing an invasive security risk to the users’ privacy. We propose a privacy-preserving implementation of single-frame method based video classification with convolutional neural networks that allows a party to infer a label from a video without necessitating the video owner to disclose their video to other entities in an unencrypted manner. Similarly, our approach removes the requirement of the classifier owner from revealing their model parameters to outside entities in plaintext. To this end, we combine existing Secure Multi-Party Computation (MPC) protocols for private image classification with our novel MPC protocols for oblivious single-frame selection and secure label aggregation across frames. The result is an end-to-end privacy-preserving video classification pipeline. We evaluate our proposed solution in an application for private human emotion recognition. Our results across a variety of security settings, spanning honest and dishonest majority configurations of the computing parties, and for both passive and active adversaries, demonstrate that videos can be classified with state-of-the-art accuracy, and without leaking sensitive user information.

Cite this Paper

BibTeX
@InProceedings{pmlr-v139-pentyala21a, title = {Privacy-Preserving Video Classification with Convolutional Neural Networks}, author = {Pentyala, Sikha and Dowsley, Rafael and De Cock, Martine}, booktitle = {Proceedings of the 38th International Conference on Machine Learning}, pages = {8487--8499}, year = {2021}, editor = {Meila, Marina and Zhang, Tong}, volume = {139}, series = {Proceedings of Machine Learning Research}, month = {18--24 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v139/pentyala21a/pentyala21a.pdf}, url = {https://proceedings.mlr.press/v139/pentyala21a.html}, abstract = {Many video classification applications require access to personal data, thereby posing an invasive security risk to the users’ privacy. We propose a privacy-preserving implementation of single-frame method based video classification with convolutional neural networks that allows a party to infer a label from a video without necessitating the video owner to disclose their video to other entities in an unencrypted manner. Similarly, our approach removes the requirement of the classifier owner from revealing their model parameters to outside entities in plaintext. To this end, we combine existing Secure Multi-Party Computation (MPC) protocols for private image classification with our novel MPC protocols for oblivious single-frame selection and secure label aggregation across frames. The result is an end-to-end privacy-preserving video classification pipeline. We evaluate our proposed solution in an application for private human emotion recognition. Our results across a variety of security settings, spanning honest and dishonest majority configurations of the computing parties, and for both passive and active adversaries, demonstrate that videos can be classified with state-of-the-art accuracy, and without leaking sensitive user information.} }
Endnote
%0 Conference Paper %T Privacy-Preserving Video Classification with Convolutional Neural Networks %A Sikha Pentyala %A Rafael Dowsley %A Martine De Cock %B Proceedings of the 38th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2021 %E Marina Meila %E Tong Zhang %F pmlr-v139-pentyala21a %I PMLR %P 8487--8499 %U https://proceedings.mlr.press/v139/pentyala21a.html %V 139 %X Many video classification applications require access to personal data, thereby posing an invasive security risk to the users’ privacy. We propose a privacy-preserving implementation of single-frame method based video classification with convolutional neural networks that allows a party to infer a label from a video without necessitating the video owner to disclose their video to other entities in an unencrypted manner. Similarly, our approach removes the requirement of the classifier owner from revealing their model parameters to outside entities in plaintext. To this end, we combine existing Secure Multi-Party Computation (MPC) protocols for private image classification with our novel MPC protocols for oblivious single-frame selection and secure label aggregation across frames. The result is an end-to-end privacy-preserving video classification pipeline. We evaluate our proposed solution in an application for private human emotion recognition. Our results across a variety of security settings, spanning honest and dishonest majority configurations of the computing parties, and for both passive and active adversaries, demonstrate that videos can be classified with state-of-the-art accuracy, and without leaking sensitive user information.
APA
Pentyala, S., Dowsley, R. & De Cock, M.. (2021). Privacy-Preserving Video Classification with Convolutional Neural Networks. Proceedings of the 38th International Conference on Machine Learning, in Proceedings of Machine Learning Research 139:8487-8499 Available from https://proceedings.mlr.press/v139/pentyala21a.html.

Related Material