Face swapping has become an emerging topic in computer vision and graphics. Indeed, many works on automatic face swapping have been proposed in recent years. These efforts have circumvented the cumbersome and tedious manual face editing processes, hence expediting the advancement in face editing. At the same time, such enabling technology has sparked legitimate concerns, particularly on its potential for being misused and abused. The popularization of "Deepfakes" on the internet has further set off alarm bells among the general public and authorities, in view of the conceivable perilous implications. Accordingly, there is a dire need for countermeasures to be in place promptly, particularly innovations that can effectively detect videos that have been manipulated.
The DeeperForensics Challenge aims at soliciting new ideas to advance the state of the art in real-world face forgery detection. The challenge uses the DeeperForensics-1.0 dataset, which is a new large-scale face forgery detection dataset proposed in CVPR 2020. DeeperForensics-1.0 represents the largest publicly available real-world face forgery detection dataset by far, with 60,000 videos constituted by a total of 17.6 million frames. Extensive real-world perturbations are applied to obtain a more challenging benchmark of larger scale and higher diversity. All source videos in DeeperForensics-1.0 are carefully collected, and fake videos are generated by a newly proposed end-to-end face swapping framework.
The dataset also features a hidden test set, which suggests a new face forgery detection setting that better simulates real-world scenarios:
Multiple sources. Fake videos in-the-wild should be manipulated by different unknown methods.
High quality. Threatening fake videos should have high quality to fool human eyes.
Diverse distortions. Different perturbations should be taken into consideration.
Thus, the hidden test set is richer in distribution than the publicly available DeeperForensics-1.0. Besides, the hidden test set will be updated constantly to get future versions along with the development of Deepfakes technology. The evaluation of the DeeperForensics Challenge is performed on the current version of hidden test set. Users are required to submit final prediction files, which we shall proceed to evaluate.