Problem
Research questions and friction points this paper is trying to address.
deepfake video detection
image-level detectors
benchmarking
video domain applicability
FakeI2V-Bench: Benchmarking the Applicability of Image-level Deepfake Detectors for Deepfake Video Detection
🤖 AI Summary
This study addresses the lack of systematic evaluation of image-level deepfake detectors in the video domain and the inadequacy of existing video-level detection benchmarks. To this end, the authors construct FakeI2V-Bench, a large-scale benchmark comprising 97,548 videos, and propose IV-Bridge, a general-purpose framework that aggregates frame-level predictions from image-level detectors using statistical features and a random forest classifier. Experimental results show that the best-performing image-level detector achieves an AUC of 80.16%, slightly outperforming current video-level methods; when enhanced by IV-Bridge, its performance improves substantially to an AUC of 93.80%, surpassing state-of-the-art video-level approaches. This work establishes a new paradigm for cross-modal deepfake detection by effectively bridging image- and video-level analysis.
📝 Abstract
Recent advances in video generation models have significantly intensified the deepfake threat, yet the current deepfake video detection benchmarks remain underdeveloped. In particular, the effectiveness of image-level detectors in the video domain has not been systematically assessed. To fill this gap, we present FakeI2V-Bench, a benchmark for evaluating state-of-the-art video-level deepfake detectors in challenging scenarios, with a particular focus on systematically assessing the performance of image-level deepfake detectors in the video domain. FakeI2V-Bench comprises 97,548 videos, containing content generated by the latest powerful generation models and covering a broader range of categories. Using this dataset, we conduct a systematic evaluation of eight video-level detectors and twelve representative image-level detectors. Experimental results show that the best-performing image-level detector achieves an 80.16% AUC, slightly outperforming the strongest video-level detector (i.e., 79.99% AUC). Going beyond benchmarking, we present IV-Bridge, a general framework that enhances the applicability of image-level deepfake detectors to videos. IV-Bridge employs a random forest model with statistical features to aggregate frame-level predictions, allowing eleven image-level detectors to surpass state-of-the-art video-level approaches, with the best-performing variant achieving a 93.80% AUC. Overall, FakeI2V-Bench establishes a rigorous benchmark for deepfake video detection and introduces a novel pathway for extending image-level detectors to the video domain, offering new insights and directions for future research. Code and data are available at https://github.com/CryptoAILab/FakeI2V-Bench.
Innovation
Methods, ideas, or system contributions that make the work stand out.
deepfake detection
image-to-video adaptation
benchmark dataset
IV-Bridge
frame-level aggregation
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
