Video Scrambling for Privacy Protection in Video Surveillance – Recent Results and Validation Framework

Frederic Dufaux (1)

1 LTCI - Laboratoire Traitement et Communication de l'Information

The issue of privacy in video surveillance has drawn a lot of interest lately. However, thorough performance analysis and validation is still lacking, especially regarding the fulfillment of privacy-related requirements. In this paper, we first review recent Privacy Enabling Technologies (PET). Next, we discuss pertinent evaluation criteria for effective privacy protection. We then put forward a framework to assess the capacity of PET solutions to hide distinguishing facial information and to conceal identity. We conduct comprehensive and rigorous experiments to evaluate the performance of face recognition algorithms applied to images altered by PET. Results show the ineffectiveness of naïve PET such as pixelization and blur. Conversely, they demonstrate the effectiveness of more sophisticated scrambling techniques to foil face recognition.

Mots clés: Video surveillance; privacy; scrambling; face recognition

Domaines: Traitement des images [eess.IV]; Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Frederic Dufaux. Video Scrambling for Privacy Protection in Video Surveillance – Recent Results and Validation Framework. Mobile Multimedia/Image Processing, Security and Applications, SPIE, Apr 2011, Orlando, United States. ⟨hal-01436403⟩