提出局部聚焦机制，提升跨域深伪检测泛化能力
A Novel Local Focusing Mechanism for Deepfake Detection Generalization
- 通过注意力机制聚焦关键局部伪造特征
- 在跨类别和跨生成模型测试中准确率提升3.7%
- 适合需要高泛化性的深伪检测应用场景
深度伪造生成技术的快速发展加剧了对鲁棒且可泛化的检测方法的需求。现有基于重建学习的方法通常依赖深层卷积网络提取差异特征，但其在不同对象类别（如人脸到汽车）和生成领域（如GAN到Stable Diffusion）间泛化性能较差，这源于深层CNN的固有局限：一是特定类别训练的模型容易过拟合语义特征分布，随网络加深而更难迁移；二是全局平均池化（GAP）将关键局部伪造线索压缩为单一向量，丢失对真实/虚假分类至关重要的判别模式。为此，我们提出一种新型局部聚焦机制（LFM），显式关注不同图像间的判别性局部特征。LFM结合显著性网络（SNet）与任务相关的Top-K池化（TKP）模块，选择K个最信息量的局部模式。为缓解Top-K池化引入的过拟合风险，引入两种正则化技术：基于排名的线性丢弃（RBLD）与随机K采样（RKS），增强模型鲁棒性。在跨域测试中，LFM相比当前最优方法邻近像素关系（NPR）实现3.7%的准确率提升和2.8%的平均精度提升，同时在单张NVIDIA A6000 GPU上保持1789 FPS的高效推理速度。该方法建立了跨域深伪检测新基准。源代码已公开于https://github.com/lmlpy/LFM.git。
原文摘要 · Abstract (English)
The rapid advancement of deepfake generation techniques has intensified the need for robust and generalizable detection methods. Existing approaches based on reconstruction learning typically leverage deep convolutional networks to extract differential features. However, these methods show poor generalization across object categories (e.g., from faces to cars) and generation domains (e.g., from GANs to Stable Diffusion), due to intrinsic limitations of deep CNNs. First, models trained on a specific category tend to overfit to semantic feature distributions, making them less transferable to other categories, especially as network depth increases. Second, Global Average Pooling (GAP) compresses critical local forgery cues into a single vector, thus discarding discriminative patterns vital for real-fake classification. To address these issues, we propose a novel Local Focus Mechanism (LFM) that explicitly attends to discriminative local features for differentiating fake from real images. LFM integrates a Salience Network (SNet) with a task-specific Top-K Pooling (TKP) module to select the K most informative local patterns. To mitigate potential overfitting introduced by Top-K pooling, we introduce two regularization techniques: Rank-Based Linear Dropout (RBLD) and Random-K Sampling (RKS), which enhance the model's robustness. LFM achieves a 3.7 improvement in accuracy and a 2.8 increase in average precision over the state-of-the-art Neighboring Pixel Relationships (NPR) method, while maintaining exceptional efficiency at 1789 FPS on a single NVIDIA A6000 GPU. Our approach sets a new benchmark for cross-domain deepfake detection. The source code are available in https://github.com/lmlpy/LFM.git
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。