Automatic Detection of Object-Based Forgery in Advanced Video

Video content authentication techniques: a comprehensive survey

“…A motion-residue-based forgery detection technique was proposed in [97], the novelty of which was that to extract forensic features from motion-residue information, it utilized feature extractors originally built for image steganalysis. The authors first used collusion operators to compute motion residues and then extracted seven steganalytic features, namely, CC-PEV, SPAM, CDF, CF, SRM, CC-JRM, and J + SRM, from these residues to model interframe and intra-frame properties of pristine (completely unmanipulated), forged, and double-compressed frames.…”

“…2.2.2. Desirable detection accuracy was reported in [97], where the proposed algorithm was tested on videos captured by static surveillance cameras. The only techniques in the literature that generated satisfactory results for realistic tampered videos taken from a standard data set SULFA [8] were [72,86,88].…”

“…Most of the videos are dedicated towards tasks, such as source camera identification. In [97], the authors tested their technique on a self-created database called SYSU-OBJFORG, although this database has not been made publically available yet. For advanced video forensic evaluation, a comprehensive collection of tampered videos similar to the image forensic databases of [145][146][147][148] is required.…”

Digital video tampering detection: An overview of passive techniques

“…performance depends on the codec used for video compression; post-processing in forgery such as tuning brightness and contrast affects the noise characteristics heavily (Pandey et al, 2014) Detection accuracy decreases with increase in compression ratio Motion residuals (Chen et al, 2015) Extact localization of forged objects in the video frame is not possible. Bit rate reduction reduces the performance of the system.…”

“…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)). …”