ABSTRACT
Citizen journalism videos increasingly complement or even replace the professional news coverage through direct reporting by event witnesses. This raises questions of the integrity and credibility of such videos. We introduce Vamos, the first user transparent video "liveness" verification solution based on video motion, that can be integrated into any mobile video capture application without requiring special user training. Vamos' algorithm not only accommodates the full range of camera movements, but also supports videos of arbitrary length. We develop strong attacks both by utilizing fully automated attackers and by employing trained human experts for creating fraudulent videos to thwart mobile video verification systems.
We introduce the concept of video motion categories to annotate the camera and user motion characteristics of arbitrary videos. We share motion annotations of YouTube citizen journalism videos and of free-form video samples that we collected through a user study. We observe that the performance of Vamos differs across video motion categories. We report the expected performance of Vamos on the real citizen journalism video chunks, by projecting on the distribution of categories. Even though Vamos is based on motion, we observe a surprising and seemingly counter-intuitive resilience against attacks performed on relatively "static" video chunks, which turn out to contain hard-to-imitate involuntary movements. We show that the accuracy of Vamos on the task of verifying whole length videos exceeds 93% against the new attacks.
- British Broadcasting Corporation. http://www.bbc.com/.Google Scholar
- Cable News Network. www.cnn.com.Google Scholar
- Citizen Evidence Lab. http://citizenevidence.org/.Google Scholar
- Citizentube. http://www.citizentube.com/.Google Scholar
- CNN iReport. Explore. Discover. Contribute. http://ireport.cnn.com/.Google Scholar
- Global Specialties R680 C-Programmable Banshi Robotic Arm. http://www.testequipmentdepot.com/products.htm?item=R680&ref=gbase&gclid=CNnOrqi2zsICFe7m7AodMS8A8A.Google Scholar
- InformaCam: Verified Mobile Media. https://guardianproject.info/apps/informacam/.Google Scholar
- Open Source Computer Vision. http://opencv.org/.Google Scholar
- RobotGeek Snapper Robotic Arm. http://www.trossenrobotics.com/robotgeek-snapper-robotic-arm?feed=Froogle&gclid=CNOOps22zsICFcxQ7Aod0GUArw.Google Scholar
- Sensor Delay. http://developer.android.com/reference/android/hardware/SensorManager.html.Google Scholar
- Stringwire. Live Video. Made Social. https://stringwire.com/.Google Scholar
- Weka. http://www.cs.waikato.ac.nz/ml/weka/.Google Scholar
- Wikipedia current events. http://en.wikipedia.org/wiki/Category:Current_events.Google Scholar
- Witness. See it. Film it. Change it. witness.org.Google Scholar
- Witness.org. http://witness.org/.Google Scholar
- YouTube. http://www.youtube.com.Google Scholar
- Youtube videos. http://seclab.cs.fiu.edu/resources.Google Scholar
- Us intelligence officials working to establish authenticity of video of sotloff being killed, reportedly by the same fighter who murdered james foley. http://www.theguardian.com/world/2014/sep/02/isis-video-steven-sotloff-beheading, September 2014.Google Scholar
- G. Abdollahian, C. M. Taskiran, Z. Pizlo, and E. J. Delp. Camera motion-based analysis of user generated video. IEEE Transactions on Multimedia, 12(1):28--41, 2010. Google ScholarDigital Library
- S. Arlot. Model selection by resampling penalization, 2007.Google Scholar
- C. M. Bishop. Neural Networks for Pattern Recognition. Oxford University Press, Inc., New York, NY, USA, 1995. Google ScholarCross Ref
- L. Breiman. Bagging predictors. Mach. Learn., 24(2):123--140, Aug. 1996. Google ScholarDigital Library
- L. Breiman. Random forests. Machine Learning, 45:5--32, 2001. Google ScholarDigital Library
- L. Cai and H. Chen. TouchLogger: Inferring Keystrokes on Touch Screen from Smartphone Motion. In Proceedings of the 6th USENIX Conference on Hot Topics in Security (HotSec), 2011. Google ScholarDigital Library
- S.-C. Chu, L. C. Jain, H.-C. Huang, and J.-S. Pan. Error-resilient triple-watermarking with multiple description coding. Journal of Networks, 5(3):267--274, 2010.Google Scholar
- R. free form dataset. http://users.cis.fiu.edu/~mrahm004/RATC/.Google Scholar
- S. I. Gallant. Perceptron-based learning algorithms. Trans. Neur. Netw., 1(2):179--191, June 1990. Google ScholarDigital Library
- P. Indyk, G. Iyengar, and N. Shivakumar. Finding pirated video sequences on the internet. Technical report, Stanford University, 1999.Google Scholar
- H. Liu, S. Saroiu, A. Wolman, and H. Raj. Software abstractions for trusted sensors. In Proceedings of the 10th International Conference on Mobile Systems, Applications, and Services, MobiSys '12, pages 365--378, 2012. Google ScholarDigital Library
- Y. Liu, H. Liu, Y. Liu, and F. Sun. User-generated-video summarization using sparse modelling. In Neural Networks (IJCNN), 2014 International Joint Conference on, pages 3909--3915, July 2014.Google ScholarCross Ref
- P. Marquardt, A. Verma, H. Carter, and P. Traynor. (sp) iPhone: Decoding Vibrations from Nearby Keyboards using Mobile Phone Accelerometers. In Proceedings of the 18th ACM Conference on Computer and Communications Security (CCS), pages 551--562. ACM, 2011. Google ScholarDigital Library
- M. MÃijller. Dynamic time warping. In Information Retrieval for Music and Motion, pages 69--84. Springer Berlin Heidelberg, 2007.Google Scholar
- M. Rahman, U. Topkara, and B. Carbunar. Seeing is Not Believing: Visual Verifications Through Liveness Analysis using Mobile Devices. In Proceedings of the 29th Annual Computer Security Applications Conference (ACSAC), 2013. Google ScholarDigital Library
- T. Shelton. The most disturbing fake videos making the rounds in Syria. http://www.globalpost.com/dispatch/news/regions/middle-east/syria/121109/fake-syria-videos-images, November 2012.Google Scholar
- W. Zhang, R. Zhang, X. Liu, C. Wu, and X. Niu. A video watermarking algorithm of h. 264/avc for content authentication. Journal of Networks, 7(8):1150--1154, 2012.Google ScholarCross Ref
Index Terms
- Liveness verifications for citizen journalism videos
Recommendations
Iconic images and citizen journalism
iConference '11: Proceedings of the 2011 iConferenceFrom the images taken by citizen photojournalists, the news media select the most emotionally engaging shots. By and large creating icons, news media uses these shots in such a way as to make them memorable. What would be otherwise mundane shots if not ...
Video Liveness for Citizen Journalism: Attacks and Defenses
The impact of citizen journalism raises important video integrity and credibility issues. In this article, we introduce Vamos, the first user transparent video “liveness” verification solution based on video motion, that accommodates the ...
Electronic governance and citizen journalism
ICEGOV '08: Proceedings of the 2nd international conference on Theory and practice of electronic governanceElectronic Governance (e-Governance) is the use of Information and Communication Technologies (ICTs) to provide citizens and organizations with more convenient access to government information and services, and to provide better public services. The ...
Comments