Preserving Privacy by De-Identifying Face Images
IEEE Transactions on Knowledge and Data Engineering
FaceTracer: A Search Engine for Large Collections of Images with Faces
ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part IV
Fully homomorphic encryption using ideal lattices
Proceedings of the forty-first annual ACM symposium on Theory of computing
Personal data vaults: a locus of control for personal data streams
Proceedings of the 6th International COnference
Negotiating privacy preferences in video surveillance systems
IEA/AIE'11 Proceedings of the 24th international conference on Industrial engineering and other applications of applied intelligent systems conference on Modern approaches in applied intelligence - Volume Part II
Discriminately decreasing discriminability with learned image filters
CVPR '12 Proceedings of the 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
The case for onloading continuous high-datarate perception to the phone
HotOS'13 Proceedings of the 14th USENIX conference on Hot Topics in Operating Systems
Hi-index | 0.00 |
We introduce SensorSift, a new theoretical scheme for balancing utility and privacy in smart sensor applications. At the heart of our contribution is an algorithm which transforms raw sensor data into a 'sifted' representation which minimizes exposure of user defined private attributes while maximally exposing application-requested public attributes. We envision multiple applications using the same platform, and requesting access to public attributes explicitly not known at the time of the platform creation. Support for future-defined public attributes, while still preserving the defined privacy of the private attributes, is a central challenge that we tackle. To evaluate our approach, we apply SensorSift to the PubFig dataset of celebrity face images, and study how well we can simultaneously hide and reveal various policy combinations of face attributes using machine classifiers. We find that as long as the public and private attributes are not significantly correlated, it is possible to generate a sifting transformation which reduces private attribute inferences to random guessing while maximally retaining classifier accuracy of public attributes relative to raw data (average PubLoss = .053 and PrivLoss = .075, see Figure 4). In addition, our sifting transformations led to consistent classification performance when evaluated using a set of five modern machine learning methods (linear SVM, kNearest Neighbors, Random Forests, kernel SVM, and Neural Nets).