通过最大化隐空间间距提升图像拼接检测的鲁棒性
Maximize margins for robust splicing detection
- 训练多个模型变体，选择隐空间间距最大的一个
- 隐空间间距越大，对后处理图像的泛化能力越强
- 适合需要高可靠性的图像取证场景
尽管拼接检测技术取得进展，基于深度学习的取证工具仍因对训练条件敏感而难以实际部署。即使对测试图像施加轻微后处理，也会显著降低检测器性能，引发其在真实场景中可靠性的担忧。本文发现，相同架构的模型在面对未见后处理时表现差异巨大，这源于训练导致的隐空间分布不同，影响样本内部的分离效果。实验表明，隐空间间距分布与模型对后处理图像的泛化能力呈强相关。据此提出实用策略：在不同条件下训练同一模型的多个变体，选择隐空间间距最大的模型，以构建更鲁棒的检测器。
原文摘要 · Abstract (English)
Despite recent progress in splicing detection, deep learning-based forensic tools remain difficult to deploy in practice due to their high sensitivity to training conditions. Even mild post-processing applied to evaluation images can significantly degrade detector performance, raising concerns about their reliability in operational contexts. In this work, we show that the same deep architecture can react very differently to unseen post-processing depending on the learned weights, despite achieving similar accuracy on in-distribution test data. This variability stems from differences in the latent spaces induced by training, which affect how samples are separated internally. Our experiments reveal a strong correlation between the distribution of latent margins and a detector's ability to generalize to post-processed images. Based on this observation, we propose a practical strategy for building more robust detectors: train several variants of the same model under different conditions, and select the one that maximizes latent margins.
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。