Lens Privacy Sealing: A New Benchmark and Method for Physical Privacy-Preserving Action Recognition

TL;DR

This paper introduces Lens Privacy Sealing (LPS), a hardware-based privacy protection method for surveillance cameras, along with a new dataset and a robust action recognition framework, achieving strong privacy and utility balance.

Contribution

The paper presents LPS as a cost-effective hardware solution for physical privacy protection, a new dataset for privacy-preserving action recognition, and a novel framework MSPNet that handles degraded videos effectively.

Findings

LPS provides superior privacy-utility trade-off compared to existing hardware methods.

MSPNet nearly doubles action recognition accuracy while maintaining low identity recognition.

LPS resists reconstruction attacks and generalizes across environments.

Abstract

RGB camera-based surveillance systems enable human action recognition for public safety and healthcare, yet raise serious privacy concerns. Existing methods rely on post-capture algorithms, which fail to protect privacy during data acquisition. We propose Lens Privacy Sealing (LPS), a simple hardware solution that physically obscures camera lenses with adjustable laminating film, providing pre-sensor privacy protection at minimal cost. Unlike software methods or expensive engineered optics, LPS achieves strong privacy through stochastic multi-layer scattering that is physically irreversible. We introduce the P$^3$AR dataset for privacy-preserving action recognition, featuring both large-scale replay-captured (P$^3$AR-NTU, 114K videos) and real-world collected (P$^3$AR-PKU) subsets with privacy attribute annotations. To handle video degradation from LPS, we propose MSPNet, a single-stage framework incorporating Inter-Frame Noise Suppressor (IFNS) and Cross-Frame Semantic Aggregator (CFSA), enhanced by contrastive language-image pre-training for robust semantic extraction. Extensive experiments demonstrate that MSPNet with IFNS and CFSA nearly doubles action recognition accuracy compared to baseline methods while suppressing identity recognition to low levels. Comprehensive validation shows LPS achieves a superior privacy-utility trade-off compared to state-of-the-art hardware methods, resists reconstruction attacks including PSF inversion and data-driven recovery, and generalizes robustly across optical configurations and challenging environments. Code is available at https://github.com/wangzy01/MSPNet.