Timezone: »

Delving into Sequential Patches for Deepfake Detection
Jiazhi Guan · Hang Zhou · Zhibin Hong · Errui Ding · Jingdong Wang · Chengbin Quan · Youjian Zhao


Recent advances in face forgery techniques produce nearly visually untraceable deepfake videos, which could be leveraged with malicious intentions. As a result, researchers have been devoted to deepfake detection. Previous studies have identified the importance of local low-level cues and temporal information in pursuit to generalize well across deepfake methods, however, they still suffer from robustness problem against post-processings. In this work, we propose the Local- & Temporal-aware Transformer-based Deepfake Detection (LTTD) framework, which adopts a local-to-global learning protocol with a particular focus on the valuable temporal information within local sequences. Specifically, we propose a Local Sequence Transformer (LST), which models the temporal consistency on sequences of restricted spatial regions, where low-level information is hierarchically enhanced with shallow layers of learned 3D filters. Based on the local temporal embeddings, we then achieve the final classification in a global contrastive way. Extensive experiments on popular datasets validate that our approach effectively spots local forgery cues and achieves state-of-the-art performance.

Author Information

Jiazhi Guan (Tsinghua University, China)
Hang Zhou (The Chinese University of Hong Kong)
Zhibin Hong (Baidu)
Errui Ding (Baidu Inc.)
Jingdong Wang (Microsoft)
Chengbin Quan (Northeastern University)
Youjian Zhao (Tsinghua University)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors