自适应路由检测伪造视频，低分辨率下仍高效准确。
Adaptive Gated Deepfake Detection for Low-Resolution and Resource-Constrained Environments
- 根据图像质量动态选择检测路径，高质图提前退出节省算力。
- 384×384分辨率下AUC达0.9708，优于现有模型且延迟低。
- 适合移动端或边缘设备部署，尤其适用于资源受限场景。
深度伪造检测模型通常依赖高质量输入、固定推理路径和高计算开销的架构，难以在低分辨率和资源受限环境中应用。本文提出AdaGate-DF，一种基于图像质量线索的自适应门控检测框架，通过双多出口结构将样本路由至不同路径：高质量图像可提前退出，节省计算资源。我们在两个基准数据集（Celeb-DF、FaceForensics++）上，对AdaGate-DF与MaD-CoRN、DefakeHop++、ShuffleNetV2在多种配置下进行评估，测试其对图像分辨率的依赖性及训练与推理效率。在Celeb-DF上，AdaGate-DF达到0.9370的AUC，优于MaD-CoRN与DefakeHop++，同时保持低推理延迟；分辨率测试显示性能随输入分辨率提升而持续改善，384×384时AUC达0.9708。FaceForensics++结果表明，即便在类别不平衡情况下，该模型仍保持良好效果，表现接近最优对比模型。总体而言，AdaGate-DF在检测性能、不确定性感知预测与计算效率之间实现了实用平衡，适用于多质量深度伪造检测场景。
原文摘要 · Abstract (English)
Deepfake detection models often rely on high-quality inputs, fixed inference paths, and computationally expensive architectures, limiting their use in low-resolution and resource-constrained settings. This paper proposes AdaGate-DF, an adaptive gated deepfake detection framework that uses image-quality cues to route samples through a dual multi-exit system so high-quality images can exit earlier and save compute. We evaluated AdaGate-DF against MaD-CoRN, DefakeHop++, and ShuffleNetV2 on two benchmark datasets (Celeb-DF and FaceForensics++) under multiple configurations to test image resolution dependence and training and inference efficiency. On Celeb-DF, AdaGate-DF achieves an AUC of 0.9370, outperforming MaD-CoRN and DefakeHop++ while maintaining a low inference latency. Resolution-based testing shows consistent improvement as input resolution increases, reaching an AUC of 0.9708 at 384 by 384. The FaceForensics++ results highlight that AdaGate-DF remains effective under class imbalance, following competitive results with evaluated models. Overall, AdaGate-DF demonstrated a practical balance between detection performance, uncertainty-aware prediction, and computational efficiency for variable-quality deepfake detection.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。