Upload
mediaeval2012
View
456
Download
2
Tags:
Embed Size (px)
DESCRIPTION
Citation preview
1
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Applying Transform-domain Scrambling to Automatically Detected Faces
Pavel Korshunov, Aleksei Triastcyn, and Touradj Ebrahimi
2
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Introduction
Recipe– Take simple face detection
(OpenCV)– Combine with a privacy filter– Apply filter to all detected regions
in a video
Focus on the privacy filter
3
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Privacy Filters
Simple: pixelization, masking, and blurring– Non-reversible– encryption anonymization, etc.
Anonymization (replacing with another object)– Non-reversible– Hard to implement
Encryption– Video alterations break the filter– Complex
4
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Transform-domain scrambling
Seed random generator with a secret key Randomly flip sign of 63 DCT coefficients
in a scrambled macro block During decoding, repeat the same
ScramblinScramblinggScramblinScramblinggTransformTransformTransformTransform Entropy Entropy
codingcodingEntropy Entropy codingcodingframe bitstream
encoder
5
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Scrambling in JPEG
F. Dufaux and T. Ebrahimi, “Scrambling for privacy protection in video surveillance systems,” IEEE Trans. on Circuits and Systems for Video Technology, vol. 18, no. 8, pp. 1168–1174, Aug 2008.
6
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Transform-domain scrambling
Pros– Reversible method– Does not negatively affect coding efficiency– Scrambling strength can be controlled– Security can be insured
Cons– Must be integrated inside the encoder
7
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Subjective evaluation results
Detection accuracy:0.24
8
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Subjective evaluation results
9
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Scrambled frame, morning video
10
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Scrambled frame, evening video
11
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Scrambled frame, evening video
12
Multimedia Signal Processing GroupSwiss Federal Institute of Technology, Lausanne
Conclusion
Face detection played unexpectedly important role
– Either use better detection or re-think evaluation methodology
Subjects were highly irritated with scrambling!
– Make scrambling more human-friendly