FaceForensics: A Large-scale Video Dataset for Forgery Detection in Human Faces

2018, Mar 24    

Andreas Rössler  Davide Cozzolino  Luisa Verdoilva  Christian Riess  Justus Thies  Matthias Nießner 

Arxiv

With recent advances in computer vision and graphics, it is now pos- sible to generate videos with extremely realistic synthetic faces, even in real time. Countless applications are possible, some of which raise a legitimate alarm, call- ing for reliable detectors of fake videos. In fact, distinguishing between original and manipulated video can be a challenge for humans and computers alike, espe- cially when the videos are compressed or have low resolution, as it often happens on social networks. Research on the detection of face manipulations has been se- riously hampered by the lack of adequate datasets. To this end, we introduce a novel face manipulation dataset of about half a million edited images (from over 1000 videos). The manipulations have been generated with a state-of-the-art face editing approach. It exceeds all existing video manipulation datasets by at least an order of magnitude. Using our new dataset, we introduce benchmarks for clas- sical image forensic tasks, including classification and segmentation, considering videos compressed at various quality levels. In addition, we introduce a bench- mark evaluation for creating indistinguishable forgeries with known ground truth; for instance with generative refinement models.

Paper Link: Arxiv 2018 - FaceForensics

Project Page: https://github.com/ondyari/FaceForensics

Project Video:

Coverage: