Obfuscation Based Privacy Preserving Representations are Recoverable Using Neighborhood Information

TL;DR

This paper demonstrates that existing geometry obfuscation techniques for privacy in visual localization are vulnerable to recovery attacks using neighborhood information, questioning their effectiveness in preserving privacy.

Contribution

The authors identify a common weakness in geometry obfuscation methods and propose a learning-based approach to recover original points, showing these schemes are not truly privacy-preserving.

Findings

Obfuscation schemes can be bypassed using neighborhood-based recovery.

Learning neighborhood descriptors enables approximate point recovery.

All tested obfuscation methods are vulnerable to the proposed attack.

Abstract

Rapid growth in the popularity of AR/VR/MR applications and cloud-based visual localization systems has given rise to an increased focus on the privacy of user content in the localization process. This privacy concern has been further escalated by the ability of deep neural networks to recover detailed images of a scene from a sparse set of 3D or 2D points and their descriptors - the so-called inversion attacks. Research on privacy-preserving localization has therefore focused on preventing these inversion attacks on both the query image keypoints and the 3D points of the scene map. To this end, several geometry obfuscation techniques that lift points to higher-dimensional spaces, i.e., lines or planes, or that swap coordinates between points % have been proposed. In this paper, we point to a common weakness of these obfuscations that allows to recover approximations of the original point positions under the assumption of known neighborhoods. We further show that these neighborhoods can be computed by learning to identify descriptors that co-occur in neighborhoods. Extensive experiments show that our approach for point recovery is practically applicable to all existing geometric obfuscation schemes. Our results show that these schemes should not be considered privacy-preserving, even though they are claimed to be privacy-preserving. Code will be available at https://github.com/kunalchelani/RecoverPointsNeighborhood.