Deluding image recognition in sift-based cbir systems

Authors:
Thanh-Toan Do;Ewa Kijak;Teddy Furon;Laurent Amsaleg
Affiliations:
Université de Rennes 1, Rennes, France;Université de Rennes 1, Rennes, France;Inria Rennes Bretagne Atlantique, Rennes, France;CNRS, Rennes, France
Venue:
Proceedings of the 2nd ACM workshop on Multimedia in forensics, security and intelligence
Year:
2010

Citing 5
Cited 5

Distinctive Image Features from Scale-Invariant Keypoints

International Journal of Computer Vision
Video copy detection: a comparative study

Proceedings of the 6th ACM international conference on Image and video retrieval
NV-Tree: An Efficient Disk-Based Index for Approximate Search in Very Large High-Dimensional Collections

IEEE Transactions on Pattern Analysis and Machine Intelligence
Secure and robust SIFT

MM '09 Proceedings of the 17th ACM international conference on Multimedia
An Image-Based Approach to Video Copy Detection With Spatio-Temporal Post-Filtering

IEEE Transactions on Multimedia

Second ACM international workshop on multimedia in forensics, security and intelligence (MiFor 2010)

Proceedings of the international conference on Multimedia
A forensic signature based on spatial distributed bag of features for image alignment and tampering detection

MiFor '11 Proceedings of the 3rd international ACM workshop on Multimedia in forensics and intelligence
Security-oriented picture-in-picture visual modifications

Proceedings of the 2nd ACM International Conference on Multimedia Retrieval
Constraint-optimized keypoint inhibition/insertion attack: security threat to scale-space image feature extraction

Proceedings of the 20th ACM international conference on Multimedia
SIFT keypoint removal and injection for countering matching-based image forensics

Proceedings of the first ACM workshop on Information hiding and multimedia security

Quantified Score

Hi-index	0.01

Visualization

Abstract

Content-Based Image Retrieval Systems used in forensics related contexts require very good image recognition capabilities. Therefore they often use the SIFT local-feature description scheme as its robustness against a large spectrum of image distortions has been assessed. In contrast, the security of SIFT is still largely unexplored. We show in this paper that it is possible to conceal images from the SIFT-based recognition process by designing very SIFT-specific attacks. The attacks that are successful in deluding the system remove keypoints and simultaneously forge new keypoints in the images to be concealed. This paper details several strategies enforcing image concealment. A copy-detection oriented experimental study using a database of 100,000 real images together with a state-of-art image search system shows these strategies are effective. This is a very serious threat against systems, endangering forensics investigations.