Saved in:
Bibliographic Details
Main Authors: Wang, Jiarui, Duan, Huiyu, Wang, Juntong, Jia, Ziheng, Yang, Woo Yi, Zhu, Xiaorong, Zhao, Yu, Qian, Jiaying, Xing, Yuke, Zhai, Guangtao, Min, Xiongkuo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.03007
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910983544373248
author Wang, Jiarui
Duan, Huiyu
Wang, Juntong
Jia, Ziheng
Yang, Woo Yi
Zhu, Xiaorong
Zhao, Yu
Qian, Jiaying
Xing, Yuke
Zhai, Guangtao
Min, Xiongkuo
author_facet Wang, Jiarui
Duan, Huiyu
Wang, Juntong
Jia, Ziheng
Yang, Woo Yi
Zhu, Xiaorong
Zhao, Yu
Qian, Jiaying
Xing, Yuke
Zhai, Guangtao
Min, Xiongkuo
contents With the rapid advancement of generative models, the realism of AI-generated images has significantly improved, posing critical challenges for verifying digital content authenticity. Current deepfake detection methods often depend on datasets with limited generation models and content diversity that fail to keep pace with the evolving complexity and increasing realism of the AI-generated content. Large multimodal models (LMMs), widely adopted in various vision tasks, have demonstrated strong zero-shot capabilities, yet their potential in deepfake detection remains largely unexplored. To bridge this gap, we present \textbf{DFBench}, a large-scale DeepFake Benchmark featuring (i) broad diversity, including 540,000 images across real, AI-edited, and AI-generated content, (ii) latest model, the fake images are generated by 12 state-of-the-art generation models, and (iii) bidirectional benchmarking and evaluating for both the detection accuracy of deepfake detectors and the evasion capability of generative models. Based on DFBench, we propose \textbf{MoA-DF}, Mixture of Agents for DeepFake detection, leveraging a combined probability strategy from multiple LMMs. MoA-DF achieves state-of-the-art performance, further proving the effectiveness of leveraging LMMs for deepfake detection. Database and codes are publicly available at https://github.com/IntMeGroup/DFBench.
format Preprint
id arxiv_https___arxiv_org_abs_2506_03007
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DFBench: Benchmarking Deepfake Image Detection Capability of Large Multimodal Models
Wang, Jiarui
Duan, Huiyu
Wang, Juntong
Jia, Ziheng
Yang, Woo Yi
Zhu, Xiaorong
Zhao, Yu
Qian, Jiaying
Xing, Yuke
Zhai, Guangtao
Min, Xiongkuo
Computer Vision and Pattern Recognition
With the rapid advancement of generative models, the realism of AI-generated images has significantly improved, posing critical challenges for verifying digital content authenticity. Current deepfake detection methods often depend on datasets with limited generation models and content diversity that fail to keep pace with the evolving complexity and increasing realism of the AI-generated content. Large multimodal models (LMMs), widely adopted in various vision tasks, have demonstrated strong zero-shot capabilities, yet their potential in deepfake detection remains largely unexplored. To bridge this gap, we present \textbf{DFBench}, a large-scale DeepFake Benchmark featuring (i) broad diversity, including 540,000 images across real, AI-edited, and AI-generated content, (ii) latest model, the fake images are generated by 12 state-of-the-art generation models, and (iii) bidirectional benchmarking and evaluating for both the detection accuracy of deepfake detectors and the evasion capability of generative models. Based on DFBench, we propose \textbf{MoA-DF}, Mixture of Agents for DeepFake detection, leveraging a combined probability strategy from multiple LMMs. MoA-DF achieves state-of-the-art performance, further proving the effectiveness of leveraging LMMs for deepfake detection. Database and codes are publicly available at https://github.com/IntMeGroup/DFBench.
title DFBench: Benchmarking Deepfake Image Detection Capability of Large Multimodal Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.03007