90%模型压缩下仍能保持检测性能，适合边缘设备部署。
A Brief Review for Compression and Transfer Learning Techniques in DeepFake Detection
- 结合剪枝、量化与知识蒸馏压缩模型，用适配器微调提升效率。
- 在同源数据上压缩90%后仍保持原性能，验证技术可行性。
- 跨模型检测时性能下降，提示需更强的领域泛化能力。
在边缘设备上训练和部署深度伪造检测模型，可通过就近处理数据来保障隐私与安全。然而，边缘端受限于计算与内存资源，难以直接运行复杂模型。为应对这一挑战，本文研究了压缩技术以降低计算开销与推理时间，并结合迁移学习减少训练成本。基于Synthbuster、RAISE和ForenSynths数据集，评估了剪枝、知识蒸馏（KD）、量化、微调及适配器（adapter-based）等方法的效果。实验表明，在训练与验证数据源自同一深度伪造模型的前提下，即使压缩率达90%，模型性能依然保持不变。但当测试数据来自训练中未见的伪造模型时，出现显著的领域泛化问题。
原文摘要 · Abstract (English)
Training and deploying deepfake detection models on edge devices offers the advantage of maintaining data privacy and confidentiality by processing it close to its source. However, this approach is constrained by the limited computational and memory resources available at the edge. To address this challenge, we explore compression techniques to reduce computational demands and inference time, alongside transfer learning methods to minimize training overhead. Using the Synthbuster, RAISE, and ForenSynths datasets, we evaluate the effectiveness of pruning, knowledge distillation (KD), quantization, fine-tuning, and adapter-based techniques. Our experimental results demonstrate that both compression and transfer learning can be effectively achieved, even with a high compression level of 90%, remaining at the same performance level when the training and validation data originate from the same DeepFake model. However, when the testing dataset is generated by DeepFake models not present in the training set, a domain generalization issue becomes evident.
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。