Fast matching for video/audio fingerprinting algorithms

Esmaeili, Mani Malek; Ward, Rabab K.; Fatourechi, Mehrdad

doi:10.1109/wifs.2011.6123134

Cited by 4 publications

(19 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• As stated in Section III-A1, the similarity matrix whose th component is often set to be inversely proportional to some dissimilarity measure . We choose the form of as in (10) rather than the exponential quadratic form as in (12): if otherwise (12) nor the linear form as in (13): if otherwise (13) (12) and (13), (10) will give the most stable partition performance. • Our proposed RAQ is given by (11).…”

Section: ) the Dimensionality Reduction Projection Inmentioning

confidence: 99%

See 1 more Smart Citation

Compact Video Fingerprinting via Structural Graphical Models

Monga

2013

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

Much previous work in video fingerprinting has focused on robustness and security issues, but the compactness requirement, i.e., the hash should be of a short length with acceptable robustness and discriminability, continues to be a significant practical challenge. In this paper, we propose a video fingerprinting method with explicit attention on compactness. First, we develop a new graphical representation of the video which reduces temporal redundancies and makes robust feature extraction much more economical. Second, a randomized adaptive quantizer is proposed to further decrease the final hash length while maintaining acceptable detection performance in terms of receiver operating characteristics (ROCs). Experimental results reveal that the proposed method offers a more favorable robustness versus discriminability tradeoff over the state of the art particularly when the bit budget of the video fingerprint is low.Index Terms-Compact video fingerprint, randomized adaptive quantizer, structural graphical models.

show abstract

Section: ) the Dimensionality Reduction Projection Inmentioning

confidence: 99%

“…This paper focuses on the first two components although fingerprint matching is also an active research area. E.g., [12] proposed a fast fingerprint matching method based on state machine which can improve the system performance. The contribution of our paper is hence twofold 1 .…”

Section: Introductionmentioning

confidence: 99%

Compact Video Fingerprinting via Structural Graphical Models

Monga

2013

IEEE Trans.Inform.Forensic Secur.

View full text Add to dashboard Cite

show abstract

“…But the complexity of computing 3D-DCT is known to be very high. We find that the DCT of temporally informative representative images (TIRI-DCT) proposed in Esmaeili et al (2011) is an appropriate feature for each sub-sequence. TIRI-DCT was originally used for a digital fingerprint for video copy detection.…”

Section: The Proposed Algorithmmentioning

confidence: 99%

“…For ω k in Formula (2-2), we choose the exponential weighting function ω γ k k = , where γ = 0 64 . according to Esmaeili et al (2011). The block size w is determined based on the results in Table 2, where the results have been obtained with different values of w. Note that, in this paper, the TPR (True Positive Rate) is calculated by dividing the number of correctly detected forgery frames by the total number of the forgery frames, and the FPR (False Positive Rate) is calculated by dividing the wrongly detected forgery frames by the total number of the no forgery frames.…”

Section: Parameters Settingsmentioning

confidence: 99%

“…The commonly used operations include format changes, signal processing operations, changes in brightness/contrast, added noise, rotation, cropping, logo insertion, compression, etc. To justify our discussions, we compare our algorithm with the correlation coefficient-based algorithm (Wang et al, 2007b), the HD-based algorithm (Lin et al, 2011) and the TIRI-DCT based algorithm (Esmaeili et al, 2011). It is worth mentioning that the TIRI-DCT based algorithm (Esmaeili et al, 2011) was originally proposed for video copy detection rather than frame duplication detection.…”

Section: Test On Video Clips Without Undergoing Video Operationsmentioning

confidence: 99%

See 1 more Smart Citation

An Improved Fingerprinting Algorithm for Detection of Video Frame Duplication Forgery

Hu¹,

Wang²,

Li³

et al.

Emerging Digital Forensics Applications for Crime Detection, Prevention, and Security

View full text Add to dashboard Cite

Frame duplication is a common way of digital video forgeries. State-of-the-art approaches of duplication detection usually suffer from heavy computational load. In this paper, the authors propose a new algorithm to detect duplicated frames based on video sub-sequence fingerprints. The fingerprints employed are extracted from the DCT coefficients of the temporally informative representative images (TIRIs) of the sub-sequences. Compared with other similar algorithms, this study focuses on improving fingerprints representing video sub-sequences and introducing a simple metric for the matching of video sub-sequences. Experimental results show that the proposed algorithm overall outperforms three related duplication forgery detection algorithms in terms of computational efficiency, detection accuracy and robustness against common video operations like compression and brightness change.

show abstract

A Fast Approximate Nearest Neighbor Search Algorithm in the Hamming Space

Esmaeili

Ward

Fatourechi

2012

IEEE Trans. Pattern Anal. Mach. Intell.

View full text Add to dashboard Cite

A fast approximate nearest neighbor search algorithm for the (binary) Hamming space is proposed. The proposed Error Weighted Hashing (EWH) algorithm is up to 20 times faster than the popular locality sensitive hashing (LSH) algorithm and works well even for large nearest neighbor distances where LSH fails. EWH significantly reduces the number of candidate nearest neighbors by weighing them based on the difference between their hash vectors. EWH can be used for multimedia retrieval and copy detection systems that are based on binary fingerprinting. On a fingerprint database with more than 1,000 videos, for a specific detection accuracy, we demonstrate that EWH is more than 10 times faster than LSH. For the same retrieval time, we show that EWH has a significantly better detection accuracy with a 15 times lower error rate.

show abstract

Fast matching for video/audio fingerprinting algorithms

Cited by 4 publications

References 7 publications

Compact Video Fingerprinting via Structural Graphical Models

Compact Video Fingerprinting via Structural Graphical Models

An Improved Fingerprinting Algorithm for Detection of Video Frame Duplication Forgery

A Fast Approximate Nearest Neighbor Search Algorithm in the Hamming Space

Contact Info

Product

Resources

About