Localization of forgeries in MPEG-2 video through GOP size and DQ analysis

Abstract

This work addresses forgery localization in MPEG-2 compressed videos. The proposed method is based on the analysis of Double Quantization (DQ) traces in frames that were encoded twice as intra (i.e., I-frames). Employing a state-of-theart method, such frames are located in the video under analysis by estimating the size of the Group Of Pictures (GOP) that was used in the first compression; then, the DQ analysis is devised for the MPEG-2 encoding scheme and applied to frames that were intra-coded in both the first and second compression. In such a way, regions that were manipulated between the two encodings are detected. Compared to existing methods based on double quantization analysis, the proposed scheme makes forgery localization possible on a wider range of settings.

Video content authentication techniques: a comprehensive survey

“…While the average frame-removal detection accuracy was 84%, for frame-addition detection, this value was 79%. The only other VPF based scheme proposed in the literature so far, [58], generated better detection results than this scheme, but only as long as the two quantization scales were carefully monitored and controlled. In addition, while [58] was applicable to MPEG-2 encoded video only, [57] was suitable for MPEG-2, 4 and H.264 encoded videos.…”

“…The only other VPF based scheme proposed in the literature so far, [58], generated better detection results than this scheme, but only as long as the two quantization scales were carefully monitored and controlled. In addition, while [58] was applicable to MPEG-2 encoded video only, [57] was suitable for MPEG-2, 4 and H.264 encoded videos. However, even though [57] was a viable method with a wide range of applicability, certain limitations need to be pointed out here.…”

“…The work in [58] addressed the issue of localizing intraframe forgeries in MPEG-2 videos. This method used the VPF based scheme proposed in [49].…”

Digital video tampering detection: An overview of passive techniques

“…Bit rate reduction reduces the performance of the system. Ghost shadow artifact (Zhang et al, 2009) Cannot accurately locate the tampered areas in each frame; works well in static background videos only Noise and quantization residue (Chetty et al, 2010;Goodwin and Chetty, 2011) Sensitive to noise, too high or too low illumination; works well static background videos only Histogram of Oriented Gradients (HOG) (Subramanyam and Emmanuel, 2012) Works with fixed GOP; copypaste tampering alone is addressed VPF, histogram of DCT coefficients (Labartino et al, 2013) Presence of B-frames are not considered; works with Variable Bit Rate (VBR) coding only Spatio-temporal coherence (Lin and Tsay, 2014) Performance decreases with increase in compression Difference between current & nontampered reference frame (Su et al, 2015a) Works with static background videos; detection accuracy decreases when the deleted foreground is very small, or too fast moving Zernike moments and 3D patch match (D'Amiano et al, 2015) Accuracy is very low M A N U S C R I P T…”

“…The works in (Milani et al, 2012b) and (Rocha et al, 2011) provide an overview on video forensics. A few works in the literature have the capability to localize the tampered regions in the video ( (Subramanyam and Emmanuel, 2012); (Labartino et al, 2013); (Lin and Tsay, 2014); (D'Amiano et al, 2015); (Feng et al, 2014)) whereas others can only classify the video as tampered or not ( (Wang and Farid, 2006); (Chen et al, 2015); (Bidokhti and Ghaemmaghami, 2015); (Su et al, 2011); (Dong et al, 2012); (Stamm et al, 2012)). …”