Visible to the public Deepfake Video Detection through Optical Flow Based CNN

TitleDeepfake Video Detection through Optical Flow Based CNN
Publication TypeConference Paper
Year of Publication2019
AuthorsAmerini, I., Galteri, L., Caldelli, R., Bimbo, A. Del
Conference Name2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)
KeywordsAI-based technologies, CNN, CNN classifiers, Computer vision, Conferences, convolutional neural nets, Deep Fakes, DeepFake, extremely realistic manipulated videos, fake information, fake video sequences, forensic technique, general opinion, Human Behavior, human factors, image classification, image sequences, Integrated optics, interframe dissimilarities, learning (artificial intelligence), Media, Metrics, multimedia contents, optical flow, optical flow fields, Optical imaging, Optical network units, Optical saturation, original video sequences, pubcrawl, public subjects, resilience, Resiliency, Scalability, synthetic videos, Video Forensics, video signal processing, visual media technology
AbstractRecent advances in visual media technology have led to new tools for processing and, above all, generating multimedia contents. In particular, modern AI-based technologies have provided easy-to-use tools to create extremely realistic manipulated videos. Such synthetic videos, named Deep Fakes, may constitute a serious threat to attack the reputation of public subjects or to address the general opinion on a certain event. According to this, being able to individuate this kind of fake information becomes fundamental. In this work, a new forensic technique able to discern between fake and original video sequences is given; unlike other state-of-the-art methods which resorts at single video frames, we propose the adoption of optical flow fields to exploit possible inter-frame dissimilarities. Such a clue is then used as feature to be learned by CNN classifiers. Preliminary results obtained on FaceForensics++ dataset highlight very promising performances.
Citation Keyamerini_deepfake_2019