Celeb-DF: A Large-scale Challenging Dataset for DeepFake Forensics

TL;DR

Celeb-DF introduces a large-scale, high-quality DeepFake video dataset to better evaluate detection algorithms, addressing limitations of previous datasets with lower visual quality and realism.

Contribution

The paper presents Celeb-DF, a new challenging dataset with 5,639 high-quality DeepFake videos, improving the resources available for DeepFake detection research.

Findings

Celeb-DF videos are of higher visual quality than previous datasets.

Existing detection methods show decreased performance on Celeb-DF.

Celeb-DF highlights the need for more robust DeepFake detection algorithms.

Abstract

AI-synthesized face-swapping videos, commonly known as DeepFakes, is an emerging problem threatening the trustworthiness of online information. The need to develop and evaluate DeepFake detection algorithms calls for large-scale datasets. However, current DeepFake datasets suffer from low visual quality and do not resemble DeepFake videos circulated on the Internet. We present a new large-scale challenging DeepFake video dataset, Celeb-DF, which contains 5,639 high-quality DeepFake videos of celebrities generated using improved synthesis process. We conduct a comprehensive evaluation of DeepFake detection methods and datasets to demonstrate the escalated level of challenges posed by Celeb-DF.