DeepfakeGenome/DeepfakeGenome

DeepfakeGenome（DFG）是一个综合性的深度伪造检测与归因基准数据集。它包含100种面部伪造算法和总计200万张图像，规模比先前的深度伪造归因（DFA）基准大4到100倍。该数据集进一步设计了4种协议进行实际评估，包括一种新颖的基于检索的归因范式。与之前的开放集评估指标不同，所提出的检索指标更符合现实世界中黑名单注册机制的主动防御情况。基于这些精心设计，该数据集研究了深度伪造归因任务的性能上限。DeepfakeGenome (DFG) is a comprehensive deepfake detection and attribution benchmark. It contains 100 facial forgery algorithms and 2M images in total, achieving 4× to 100× larger than prior DFA benchmarks. The dataset further designed 4 protocols for practical evaluation, including a novel retrieval-based attribution paradigm. Unlike previous open-set evaluation metrics, the proposed retrieval metrics are more aligned with the real-world active defense situation of blacklist registration mechanisms. Based on these elaborate designs, the dataset investigates the performance ceiling of deepfake attribution task.