On 1 January 2016 formally, and at the end of the month practically, a new research project in which I am involved started. It is called InVID, deals with the algorithm-supported verification of user-generated video, and runs from 2016 until the end of 2018. It looks promising, and can largely build on what we have been doing in REVEAL. The main difference though: InVID will focus on video verification, something REVEAl did not tackle. Hence an ideal continuation in my verification activities and dealing with eyewitness media.
Here’s some more information on the project (an edited and abriged version of this text has also been published on the DW Innovation blog).
InVID (In Video Veritas) aims to assist in and partly automate the verification of videos captured by eyewitnesses, also referred to as user-generated content. A main goal is to make such videos usable for news providers for inclusion in their reporting, situation assessment, publication and the like.
Nowadays news organizations often rely on user generated recordings of (breaking) news events. However, the footage might have been manipulated in many ways (e.g. image manipulations, misleading dates or locations). News providers are confronted with the fundamental need to verify such contents and deal with it appropriately. Some of the challenges that need to be dealt with are the verification itself, appropriate rights management, source protection and the like.
The EC co-funded project InVID intends to develop a platform that provides services and features that detect, authenticate and check the reliability and accuracy of newsworthy video files spread via social networks. The InVID platform is expected to allow broadcasters, news agencies, web-only players, newspapers and publishers to integrate social media video content into their news output in a faster and more efficient way, aiding them in the verification process.
InVID (In Video Veritas – Verification of Social Media Video Content for the News Industry) is an Innovation Action that has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 687786. InVID started in January 2016 and will run until December 2018. Its goal is to build a platform providing services to detect, authenticate and check the reliability and accuracy of newsworthy video files and video content spread via social media. The project is coordinated by CERTH (Centre for Research and Technology Hellas, Greece), and the consortium’s partners are MODUL Technology GmbH (Austria), Universitat de Lleida (Spain), EXO MAKINA (France), webLyzard technology GmbH (Austria), Condat AG (Germany), APA-IT Informations Technologie GmbH (Austria), Agence France Presse (France), and Deutsche Welle (Germany).