发表机构
Kansas State University（堪萨斯州立大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过场景识别实验证实，CNNs对真实与虚假图像的隐藏层激活存在可统计的差异，该差异无法仅用图像退化解释，为改进虚假图像检测提供了新思路。
AI 中文摘要
虚假/合成图像日益普遍，但卷积神经网络（CNNs）是否以相同的内部方式处理真实图像与虚假图像仍不明确。本研究检验CNNs对真实图像与虚假图像的表示存在差异这一假设，即即便语义内容保持一致，虚假图像也会诱导不同的隐藏层激活模式。该假设在场景识别场景中使用训练好的CNN模型进行评估：提取密集层激活，采用神经符号方法为选定神经元分配语义标签；针对每张真实测试图像，使用基于Stable Diffusion变体的物体标签引导文本到图像、图像到图像生成技术，生成语义内容相似的对应虚假图像，随后对配对的真实-虚假激活模式进行统计比较。此外，还通过另一数据集、CNN架构、生成模型的额外实验，以及JPEG/模糊退化分析来评估结果的鲁棒性。结果表明，虚假图像会引发不同的隐藏神经元激活，且这些差异无法仅用简单的图像退化来解释。总体而言，研究发现至少在部分场景下，真实图像与虚假图像在CNN隐藏层激活行为上存在差异，这为利用该差异改进虚假图像检测的后续研究开辟了方向。
英文摘要
Fake/synthetic images are increasingly prevalent, but it remains unclear whether Convolutional Neural Networks (CNNs) process real and fake images in the same internal manner. This work examines the hypothesis that CNNs represent real and fake images differently, such that fake images induce different hidden-layer activation patterns even when semantic content is preserved. The hypothesis is evaluated in scene recognition settings using trained CNN models. Dense-layer activations are extracted, and neurosymbolic methods assign semantic labels to selected neurons. For each real test image, corresponding fake images are generated with similar semantic content using object-label-guided text-to-image and image-to-image generation based on Stable Diffusion variants. Paired real-fake activation patterns are then compared statistically. Additional experiments with another dataset, CNN architecture, generative model, and JPEG/blur degradation analysis assess robustness. Results suggest that fake images evoke different hidden-neuron activations, and these differences are not explained only by simple image degradation. Overall, the findings indicate that real and fake images differ in CNN hidden-layer activation behavior at least in some settings, which opens the door for follow-up work on making use of this different behavior to improve fake image detection.
