统一 KANs 的局部性与特征漂移补偿投影的无数据重放持续人脸伪造检测
Unifying Locality of KANs and Feature Drift Compensation Projection for Data-free Replay based Continual Face Forgery Detection
AI总结：
针对持续人脸伪造检测中的灾难性遗忘问题，提出基于 KAN 的 KAN-CFD 框架，通过 DG-KD 保留局部性以拟合高维图像，并利用 FS-KDCP 在无数据重放下避免输入空间重叠，显著提升检测性能并减少遗忘。
摘要：
人脸伪造技术的快速发展要求检测器不断适应新的伪造方法，从而将人脸伪造检测置于持续学习范式中。然而，当检测器学习新的伪造类型时，它们在先前类型上的性能通常会迅速下降，这种现象被称为灾难性遗忘。Kolmogorov-Arnold Networks (KANs) 利用局部可塑样条作为激活函数，使其能够通过仅修改函数的局部区域来学习新任务，同时保持其他区域不受影响。因此，它们天然适合解决灾难性遗忘。然而，KANs 有两个显著的局限性：1) 样条对高维图像建模效果不佳，而适用于图像的替代激活函数缺乏局部性这一关键属性；2) 在持续学习中，当来自不同域的特征发生重叠时，由于对相同区域的重复修改，将不同域映射到不同曲线区域的过程总是会发生崩溃。在本文中，我们提出了一种基于 KAN 的持续人脸伪造检测（KAN-CFD）框架，其中包含 Domain-Group KAN Detector (DG-KD) 和基于 KAN 漂移补偿投影的无数据重放特征分离策略（FS-KDCP）。DG-KD 使 KANs 能够拟合高维图像输入，同时保留局部性和局部可塑性。FS-KDCP 在不使用先前任务数据的情况下，避免了 KAN 输入空间的重叠。实验结果表明，所提出的方法实现了卓越的性能，同时显著减少了遗忘。
英文摘要：
The rapid advancements in face forgery techniques necessitate that detectors continuously adapt to new forgery methods, thus situating face forgery detection within a continual learning paradigm. However, when detectors learn new forgery types, their performance on previous types often degrades rapidly, a phenomenon known as catastrophic forgetting. Kolmogorov-Arnold Networks (KANs) utilize locally plastic splines as their activation functions, enabling them to learn new tasks by modifying only local regions of the functions while leaving other areas unaffected. Therefore, they are naturally suitable for addressing catastrophic forgetting. However, KANs have two significant limitations: 1) the splines are ineffective for modeling high-dimensional images, while alternative activation functions that are suitable for images lack the essential property of locality; 2) in continual learning, when features from different domains overlap, the mapping of different domains to distinct curve regions always collapses due to repeated modifications of the same regions. In this paper, we propose a KAN-based Continual Face Forgery Detection (KAN-CFD) framework, which includes a Domain-Group KAN Detector (DG-KD) and a data-free replay Feature Separation strategy via KAN Drift Compensation Projection (FS-KDCP). DG-KD enables KANs to fit high-dimensional image inputs while preserving locality and local plasticity. FS-KDCP avoids the overlap of the KAN input spaces without using data from prior tasks. Experimental results demonstrate that the proposed method achieves superior performance while notably reducing forgetting.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
