Strategies for Countering Fake Information: new trends in multimedia authenticity verification and source identification
Nowadays, with the prevalence of low-cost imaging devices (smartphones, tablets, camcorders, digital cameras, scanners, wearable and IoT devices), images and videos have become the main modalities of information being exchanged in everyday life. The ever-increasing convenience of image acquisition has facilitated instant distribution and sharing of multimedia on digital social platforms. In the meantime, powerful media editing tools allow people to easily manipulate digital content for malicious or criminal ends, up to create completely new and photorealistic image and video with the use of sophisticated AI techniques. In response to this threat, multimedia forensics community has produced a major research effort regarding source identification and image forgery detection. In fact, in all cases where multimedia serves as critical evidence, forensic technologies that help to determine the origin, the authenticity of sources and integrity of multimedia content become essential in forensic investigations or in fake news debunking.
In detail, in this seminar forensic methodologies and tools employed so far to analyze digital evidences to ensure media authenticity will be outlined with special attention to the most promising solutions relying on Convolutional Neural Networks; an in-depth focus will be dedicated to realistic scenarios, such as the spreading of multimedia data over social networks. Finally, an overview of the recent trends and evolution will be provided giving an insight on new security issues related to deepfakes and computer generated images with GAN.