D题参考资料

2744557306

参考资料一
: G5 }' W1 i  a0 k  N' }

人工智能生成的图像与真实图像的对比度相比，纹理斑块丰富和贫乏的对比度要高得多。这样我们用肉眼就可以看到区别了，所以可以将对比度的结果放入可训练模型，并将结果数据输入分类器，这样就是我们这篇论文的模型架构：

" L# o8 P4 ?7 o参考资料二

This paper explores the task of detecting images generated by text-to-image diffusion models. To evaluate this, we consider images generated from captions in the MSCOCO and Wikimedia datasets using two state-of-the-art models: Stable Diffusion and GLIDE. Our experiments show that it is possible to detect the generated images using simple Multi-Layer Perceptrons (MLPs), starting from features extracted by CLIP, or traditional Convolutional Neural Networks (CNNs). We also observe that models trained on images generated by Stable Diffusion can detect images generated by GLIDE relatively well, however, the reverse is not true. Lastly, we find that incorporating the associated textual information with the images rarely leads to significant improvement in detection results but that the type of subject depicted in the image can have a significant impact on performance. This work provides insights into the feasibility of detecting generated images, and has implications for security and privacy concerns in real-world applications.
+ u# K! f1 _* W& q/ V/ m% Y# p2 D本文探讨了检测由文本到图像扩散模型生成的图像的任务。为了评估这一点，我们使用两种最先进的模型:稳定扩散和滑翔，考虑从MSCOCO和维基媒体数据集中的标题生成的图像。我们的实验表明，可以使用简单的多层感知器(mlp)从CLIP提取的特征或传统的卷积神经网络(cnn)开始检测生成的图像。[backcolor=rgba(255, 228, 181, 0.7)]我们还观察到，在Stable Diffusion生成的图像上训练的模型可以较好地检测GLIDE生成的图像，反之则不然。最后，我们发现将相关的文本信息与图像结合很少会导致检测结果的显着改善，但是图像中描述的主题类型会对性能产生显着影响。这项工作为检测生成图像的可行性提供了见解，并对现实应用中的安全和隐私问题产生了影响。
" p3 ^! G! A- s1 u  K1 F
为大家提供了项目，以及项目的论文，大家可以参考一下

参考文献三：
关于扩散模型生成的合成图像的检测项目地点：GitHub - grip-unina/DMimageDetection: On the detection of synthetic images generated by diffusion models
参考文献四：
用于扩散生成图像检测的 DIRE
7 f" R( ]- P' W  ^项目地址
8 |$ q1 F8 ]4 G# S+ K# y" Xhttps://github.com/ZhendongWang6/DIRE?tab=readme-ov-file

4 X5 x& {! H! r* s% F

hyd1126

很有帮助