🧪 Test?View on arXiv
Lensless Gaze Is Not Private by Default: Auditing Identity Leakage Across Disclosure Surfaces
Not specified in the provided content
privacyidentity leakagesensingdata security
2609.09188
Builder Relevance
1h ago70%
Abstract
The paper argues that lensless near-eye sensing, often considered privacy-friendly, can still leak identity information when evaluated under specific conditions.
Reality Card
Core Claim
The study demonstrates that privacy claims for lensless sensing must be empirically tested at disclosure boundaries, revealing significant identity leakage through various representations.
Method / Result
Simulated lensless measurements yield 96.7% top-1 identification accuracy.
Limitations
The study focuses on a known-gallery closed-set identification protocol, which may not generalize to unknown or varying optical keys.
Paper to code
Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.
No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.