ECCV 2026 Paper: Privacy Leakage in Scene Coordinate Regression Models

CSProfKGD · x · 2026-08-06

An upcoming ECCV 2026 paper (Long Oral) exposes privacy vulnerabilities in neural scene representations.

Scene Coordinate Regression (SCR) models are traditionally considered privacy-preserving because scenes are implicitly encoded within network weights rather than stored explicitly as images or maps. However, the authors demonstrate a query-based attack that breaks this assumption.

Experiments on indoor and outdoor datasets confirm that sensitive training environments can be reconstructed with high fidelity.

Original post →

More from Safety

Safety channel →