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