TY - GEN
T1 - Privid
T2 - 19th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2022
AU - Cangialosi, Frank
AU - Agarwal, Neil
AU - Arun, Venkat
AU - Jiang, Junchen
AU - Narayana, Srinivas
AU - Sarwate, Anand
AU - Netravali, Ravi
N1 - Publisher Copyright:
© 2022 by The USENIX Association. All Rights Reserved.
PY - 2022
Y1 - 2022
N2 - Analytics on video recorded by cameras in public areas have the potential to fuel many exciting applications, but also pose the risk of intruding on individuals' privacy. Unfortunately, existing solutions fail to practically resolve this tension between utility and privacy, relying on perfect detection of all private information in each video frame-an elusive requirement. This paper presents: (1) a new notion of differential privacy (DP) for video analytics, (ρ,K,ǫ)-eventduration privacy, which protects all private information visible for less than a particular duration, rather than relying on perfect detections of that information, and (2) a practical system called Privid that enforces duration-based privacy even with the (untrusted) analyst-provided deep neural networks that are commonplace for video analytics today. Across a variety of videos and queries, we show that Privid increases error by 1-21% relative to a non-private system.
AB - Analytics on video recorded by cameras in public areas have the potential to fuel many exciting applications, but also pose the risk of intruding on individuals' privacy. Unfortunately, existing solutions fail to practically resolve this tension between utility and privacy, relying on perfect detection of all private information in each video frame-an elusive requirement. This paper presents: (1) a new notion of differential privacy (DP) for video analytics, (ρ,K,ǫ)-eventduration privacy, which protects all private information visible for less than a particular duration, rather than relying on perfect detections of that information, and (2) a practical system called Privid that enforces duration-based privacy even with the (untrusted) analyst-provided deep neural networks that are commonplace for video analytics today. Across a variety of videos and queries, we show that Privid increases error by 1-21% relative to a non-private system.
UR - http://www.scopus.com/inward/record.url?scp=85132423815&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85132423815&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85132423815
T3 - Proceedings of the 19th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2022
SP - 209
EP - 229
BT - Proceedings of the 19th USENIX Symposium on Networked Systems Design and Implementation, NSDI 2022
PB - USENIX Association
Y2 - 4 April 2022 through 6 April 2022
ER -