---
title: "Dynamic Inconsistency-aware DeepFake Video Detection"
authors:
  - "Ziheng Hu"
  - "Hongtao Xie"
  - "YuXin Wang"
  - "Jiahong Li"
  - "Zhongyuan Wang"
  - "Yongdong Zhang"
doi: "10.24963/ijcai.2021/102"
url: "https://ijcai.org/proceedings/2021/102"
pdf: "https://ijcai.org/proceedings/2021/0102.pdf"
venue: "Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence"
year: 2021
pages: "736-742"
track: "Main Track"
keywords:
  - "Computer Vision: Biometrics, Face and Gesture Recognition"
  - "Computer Vision: Recognition: Detection, Categorization, Indexing, Matching, Retrieval, Semantic Interpretation"
---

# Dynamic Inconsistency-aware DeepFake Video Detection

Ziheng Hu, Hongtao Xie, YuXin Wang, Jiahong Li, Zhongyuan Wang, Yongdong Zhang

*Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence*, pages 736-742. DOI: [10.24963/ijcai.2021/102](https://doi.org/10.24963/ijcai.2021/102)

## Abstract

The spread of DeepFake videos causes a serious threat to information security, calling for effective detection methods to distinguish them. However, the performance of recent frame-based detection methods become limited due to their ignorance of the inter-frame inconsistency of fake videos. In this paper, we propose a novel Dynamic Inconsistency-aware Network to handle the inconsistent problem, which uses a Cross-Reference module (CRM) to capture both the global and local inter-frame inconsistencies. The CRM contains two parallel branches. The first branch takes faces from adjacent frames as input, and calculates a structure similarity map for a global inconsistency representation. The second branch only focuses on the inter-frame variation of independent critical regions, which captures the local inconsistency. To the best of our knowledge, this is the first work to totally use the inter-frame inconsistency information from the global and local perspectives. Compared with existing methods, our model provides a more accurate and robust detection on FaceForensics++, DFDC-preview and Celeb-DFv2 datasets.