VISION: a video and image dataset for source identification

Abstract

Forensic research community keeps proposing new techniques to analyze digital images and videos. However, the performance of proposed tools are usually tested on data that are far from reality in terms of resolution, source device, and processing history. Remarkably, in the latest years, portable devices became the preferred means to capture images and videos, and contents are commonly shared through social media platforms (SMPs, for example, Facebook, YouTube, etc.). These facts pose new challenges to the forensic community: for example, most modern cameras feature digital stabilization, that is proved to severely hinder the performance of video source identification technologies; moreover, the strong re-compression enforced by SMPs during upload threatens the reliability of multimedia forensic tools. On the other hand, portable devices capture both images and videos with the same sensor, opening new forensic opportunities. The goal of this paper is to propose the VISION dataset as a contribution to the development of multimedia forensics. The VISION dataset is currently composed by 34,427 images and 1914 videos, both in the native format and in their social version (Facebook, YouTube, and WhatsApp are considered), from 35 portable devices of 11 major brands. VISION can be exploited as benchmark for the exhaustive evaluation of several image and video forensic tools.

Combining PRNU and noiseprint for robust and efficient device source identification

“…In addition, a good performance on small regions allows one to use this approach for image forgery detection and localization, especially the critical case of small-size forgeries. In Fig.3, we show the histograms of the PRNU-based (left) and noiseprint-based (right) distances computed on the widespread VISION dataset [23] in various situations (different rows). Each subplot shows three different histograms:…”

PRNU-based source device attribution for YouTube videos

“…To evaluate the efficiency of the proposed methods, we used a public dataset (VISION) [9] which comprises native and social media videos. In the dataset, were extracted from videos using ffprobe, included in ffmpeg software.…”

“…Some authors like Samet et al in [8] just assume that I frames are the best to be used, others like Dasara et al in [9] give equal importance to I, P and B frames, and use all video frames for fingerprint estimation. Accordingly, they reported that a low accuracy in source attribution is obtained when performed on videos re-compressed by YouTube or Whatsapp (compared to their native version).…”

“…We call this method "block-based" approach since it relies on block-wise inframe noise analysis. The proposed and frame-based methods are tested with a wide range of videos available in the VISION database [9] acquired from various smart-phones and tablets ( Table 2).…”

Analysis of Rolling Shutter Effect on ENF-Based Video Forensics

“…Camera forensics has become a significant field of research for various applications such as verification query image/video pairs can be attributed to the same source camera, and verification the source of a suspicious image/video [23], [24]. To the best of our knowledge, the only work exploiting ENF for camera forensics is that of Hajj-Ahmad et al [18].…”