针对AIGC检测的鲁棒性测试——常见攻击手段汇总

前言:这篇文章来总结一下针对AIGC检测的常见攻击手段,选取的研究工作均出自近5年AIGC检测相关文章。(论文被拒了需要补实验,先来看看别人怎么做的……)

2019 WIFS Detecting and Simulating Artifacts in GAN Fake Images

We show the robustness of the proposed method with two different post-processing methods: JPEG compression and image resize. For JPEG compression, we randomly select one JPEG quality factor from [100, 90, 70, 50] and apply it to each of the fake image. For image resize, we randomly select one image size from [256, 200, 150, 128].

在这里插入图片描述

2020 CVPR CNN-Generated Images Are Surprisingly Easy to Spot… for Now

To test this, we blurred (simulating re-sampling) and JPEG-compressed the real and fake images following the protocol in [38], and evaluated our ability to detect them (Figure 5).

在这里插入图片描述

2022 ECCV Detecting Generated Images by Real Images

Gaussian blurring (sigma: 0.1–1), JPEG quality factors (70–100), image cropping, and resizing (cropping/scaling factor: 0.25–1)

在这里插入图片描述

2022 ICIP Fusing Global and Local Features for Generalized AI-Synthesized Image Detection

We also evaluate the robustness of our method and the baselines on the images post-processed with Gaussian Blur and JPEG Compression.

在这里插入图片描述

2023 APSIPA ASC AI-Generated Image Detection using a Cross-Attention Enhanced Dual-Stream Network

In this section, we assess the robustness of our proposed model in the face of seven post-processing techniques, encompassing adjustments to chromaticity, brightness, contrast, sharpness, rotation, and the application of Gaussian blur and mean blur… To create a more realistic simulation of complex real-world scenarios, we’ve incorporated randomness into the parameters controlling the image alterations. For instance, the factors governing the degree of image manipulation (chromaticity, brightness, contrast) are randomly selected from a range of 0.5 to 2.5 for each image in the test dataset. Similarly, the factor controlling image sharpness is an arbitrary integer within the range of 0 to 4. Rotation degrees range from 0 to 360, and the kernel size for both Gaussian and mean filters is 5 × 5.

在这里插入图片描述

2023 ICASSP On the detection of synthetic images generated by diffusion models

For each image of the test, a crop with random (large) size and position is selected, resized to 200 × 200 pixels, and compressed using a random JPEG quality factor from 65 to 100.

在这里插入图片描述

2023 CVPR Learning on Gradients- Generalized Artifacts Representation for GAN-Generated Images Detection

To evaluate the robustness of the proposed framework to image perturbations, we apply common image perturbations on the test images with a probability of 50% following [13]. These perturbations include blurring, cropping, compression, adding random noise, and a combination of all of them. In this subsection, the discriminator of StyleGANbedroom is used as the transformation model.

在这里插入图片描述

2023 CCS DE-FAKE- Detection and Attribution of Fake Images Generated by Text-to-Image Generation Models

Specifically, we evaluate the robustness of detectors and attributors against adversary example attacks, which are the most common and severe attacks against machine learning models. We leverage three representative adversary example attacks, namely FGSM [14], BIM [18], and DI-FGSM [41] to conduct the robustness analysis. Furthermore, given that our hybrid detector and attributor consider both the image and its corresponding prompt, we propose HybridFool, which maximizes the distance between the embedding of a given image and the prompt by adding perturbations to the image. In the following, we first present each adversary example attack we consider in this robustness analysis. Then, we show the evaluation results.

在这里插入图片描述

2023 ICCV DIRE for Diffusion-Generated Image Detection

Here, we evaluate the robustness of detectors in two-class degradations, i.e., Gaussian blur and JPEG compression, following [47]. The perturbations are added under three levels for Gaussian blur (σ = 1, 2, 3) and two levels for JPEG compression (quality = 65, 30).

在这里插入图片描述

2024 CVPR-W Raising the Bar of AI-generated Image Detection with CLIP

JPEG compression (100-60), WEBP compression (100-60), Resizing (1.25, 1.0, 0.75, 0.5, 0.25)
在这里插入图片描述

2024 CVPR AEROBLADE- Training-Free Detection of Latent Diffusion Images Using Autoencoder Reconstruction Error

Following previous works [12, 59] we use JPEG compression (with quality q), center cropping (with crop factor f and subsequent resizing to the original size), Gaussian blur, and Gaussian noise (both with standard deviation σ).

在这里插入图片描述

2024 CVPR Forgery-aware Adaptive Transformer for Generalizable Synthetic Image Detection

Specifically, we adopt random cropping, Gaussian blurring, JPEG compression, and Gaussian noising, each with a probability of 50%.

在这里插入图片描述

2024 ICML DRCT- Diffusion Reconstruction Contrastive Training towards Universal Detection of Diffusion Generated Images

we adopt the experimental setup from (Wu et al., 2023a) to perform resizing (with scales of 0.5, 0.75, 1.0, 1.25, 1.5) and JPEG compression (with quality factors of 60, 70, 80, 90, 100) on the tested images, which include both real and generated images

在这里插入图片描述

20231102 arXiv Detecting Generated Images by Real Images Only

The post-processing operations we used include:
• Blurring: Gaussian filtering with a kernel size of 3 and sigma from 0.1 to 1.
• Brightness adjustment: the adjustment parameter is from 0.3 to 3.
• Contrast adjustment: Gamma transform with γ from 0.3 to 3.
• Random cropping: for a 256×256 image, the cropping size was from 256 to 96, and we up-sampled the amplitude spectrum of the LNP of the final cropped image back to 256.
• JPEG compression: quality factors from 70 to 100.
• Gaussian noise: the sigma was set from 1 to 10, and the PSNR of the original image and the image after adding noise is from 26 to 47.
• Pepper & Salt noise: the ratio of pepper to salt is 1:1. The density of the added noise is 0.001 to 0.01, and the PSNR of the noisy images is from 18 to 31.
• Speckle noise: the sigma ranges from 0.01 to 0.1, and the PSNR of the noisy images is from 22 to 57.
• Poisson noise: the lambda is set from 0.1 to 1, and the PSNR of the noisy images is from 3 to 58.

在这里插入图片描述

20240325 arXiv Let Real Images be as a Judger, Spotting Fake Images Synthesized with Generative Models

blur (0-3.0), compression (100-30), noise (0-3.0), resizing (1.0-0.2)
在这里插入图片描述

无鲁棒性检测实验(极个别现象)

  • 2023 CVPR Towards Universal Fake Image Detectors that Generalize Across Generative Models
  • 2024 CVPR LaRE2 Latent Reconstruction Error Based Method for Diffusion-Generated Image Detection

总结:常见的针对AIGC检测的鲁棒性测试手段有:JPEG压缩、resize(图像尺寸调整)、高斯模糊、高斯噪声,还有一些零散的攻击手段如图像裁剪、色度、亮度、对比度、锐化、旋转、随机噪声、椒盐噪声、散斑噪声、泊松噪声、对抗样本……注意在测试每种攻击手段时,要把攻击程度考虑进去,比如JPEG压缩就要考虑压缩因子。

本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如若转载,请注明出处:http://www.mzph.cn/diannao/34514.shtml

如若内容造成侵权/违法违规/事实不符,请联系多彩编程网进行投诉反馈email:809451989@qq.com,一经查实,立即删除!

相关文章

JavaScript的学习之事件的简介

目录 一、事件是什么 二、如何处理事件 一、事件是什么 定义:事件就是浏览器和用户之间的交互行为。 例如:点击按钮、鼠标移动、关闭窗口等。 二、如何处理事件 我们可以在对应的事件属性中设置一些JS行为,当事件触发的时候会将这些代码执行…

关于Pandas的数据填充

前言 在数据分析与预处理过程中,脏数据几乎不可避免,这直接影响到后续分析的准确性和可靠性。清洗数据中最常见的就是处理空值。Pandas DF的数据填充功能非常强大。本文介绍Pandas中常用的几种数据填充(也称为缺失值处理)方法&am…

35岁,是终点?还是拐点?

35岁,是终点还是拐点,取决于我们对生活和事业的态度、目标以及行动。这个年龄可以看作是一个重要的转折点,具有多重意义和可能性。 很多人在35岁时,已经在自己的职业生涯中建立了一定的基础,可能达到了管理层或专家级别…

Python量化交易学习——Part8:带有技术因子指标的多因子策略

技术面分析又称技术分析(Technical Analysis ),是股票投资分析的专业术语。技术分析研究以往价格和交易量数据,进而预测未来的价格走向。此类型分析侧重于图表与公式的构成,以捕获主要和次要的趋势,并通过估测市场周期长短,识别买入 / 卖出机会。根据您选择的时间跨度,…

Charles抓包工具系列文章(二)-- Repeat 回放http请求

一、什么是http请求回放 当我们对客户端进行抓包,经常会想要重试http请求,或者改写原有部分进行重新请求,都需要用到回放http请求。 还有一种场景是压力测试,对一个请求进行重复请求多少次,并加上适当的并发度。 这里…

【PythonWeb开发】Flask视图函数传递数据到前端模版的方法总结。

在Flask框架中,视图函数返回响应有四种常见方式,都得掌握。 一、返回文本内容 可以直接返回字符串,Flask会自动将其转换为一个响应对象,具有默认的text/html内容类型。 app.route(/return_text) def return_text():return "…

力扣第211题“添加与搜索单词 - 数据结构设计”

关注微信公众号 数据分析螺丝钉 免费领取价值万元的python/java/商业分析/数据结构与算法学习资料 在本篇文章中,我们将详细解读力扣第211题“添加与搜索单词 - 数据结构设计”。通过学习本篇文章,读者将掌握如何实现一个支持通配符搜索的字典数据结构&…

WebStorm 配置 PlantUML

1. 安装 PlantUML 插件 在 WebStorm 插件市场搜索 PlantUML Integration 并安装,重启 WebStorm 使插件生效。 2. 安装 Graphviz PlantUML 需要 Graphviz 来生成图形。使用 Homebrew 安装 Graphviz: 打开终端(Terminal)。确保你…

java的Nio演进

提示:文章写完后,目录可以自动生成,如何生成可参考右边的帮助文档 java的Nio演进 前言简介正文总结 前言 在JDK1.4推出JavaNIO之前,基于Java的所有Socket通信都采用了同步阻塞模式 (BIO),这种一请求一应答的通信模型简化了上层的应用开发,但…

C语言小例程28/100

题目&#xff1a;利用递归方法求5!。 程序分析&#xff1a;递归公式&#xff1a;fnfn_1*4! #include <stdio.h>int main() {int i;int fact(int);for(i0;i<6;i){printf("%d!%d\n",i,fact(i));} } int fact(int j) {int sum;if(j0){sum1;} else {sumj*fac…

消防认证-防火门

一、消防认证 消防认证是指消防产品符合国家相关技术要求和标准&#xff0c;且通过了国家认证认可监督管理委员会审批&#xff0c;获得消防认证资质的认证机构颁发的证书&#xff0c;消防产品具有完好的防火功能&#xff0c;是住房和城乡建设领域验收的重要指标。 二、认证依…

嵌入式系统中串口通信粘包问题的解决方案(C语言)

文章目录 0. 引言1. 什么是粘包问题&#xff1f;2. 粘包问题的影响3. 处理粘包问题的思路4. 不同处理方法的优缺点分析5. 实现方案5.1 数据包格式5.2 代码实现 0. 引言 在嵌入式系统中&#xff0c;串口通信是一种常见且重要的数据传输方式。然而&#xff0c;由于硬件和软件的限…

闪迪sd卡视频格式化数据恢复方法,你了解吗

咨询&#xff1a;“我不小心将闪迪SD卡格式化了&#xff0c;里面的重要视频文件全都不见了。我感到非常焦虑&#xff0c;因为这些视频对我来说意义非凡。现在急需找到方法来恢复&#xff01;&#xff01;” 在数字时代&#xff0c;SD卡已成为我们日常生活中不可或缺的数据存储设…

python中数据的作用域

一、命名空间 在 Python 中&#xff0c;命名空间是一个系统&#xff0c;它用于确保名字的唯一性&#xff0c;并防止命名冲突。命名空间是一个存储变量名称&#xff08;或者更广泛地说&#xff0c;标识符&#xff09;与对象之间映射的抽象概念。每个变量名你在程序中创建&#x…

本篇内容:ArkTS开发系列之事件(2.8.1触屏、键鼠、焦点事件)

上篇回顾&#xff1a; ArkTS开发系列之导航 (2.7动画&#xff09; 本篇内容&#xff1a;ArkTS开发系列之事件&#xff08;2.8.1触屏、键鼠、焦点事件&#xff09; 一、知识储备 1. 触屏事件&#xff1a;包括点击事件、拖拽事件、触摸事件。 点击事件 Button()....onClick(…

msvcp120.dll丢失怎么办,找不到msvcp120.dll的多种解决方法

最近&#xff0c;我在运行一个程序时遇到了一个错误&#xff0c;系统提示找不到msvcp120.dll文件&#xff0c;无法继续执行代码。这让我感到非常困扰&#xff0c;因为这个问题导致我无法正常运行这个程序。经过一番搜索和尝试&#xff0c;我找到了几种修复这个问题的方法&#…

如何开发、使用 Starter

开发 第一步&#xff1a;创建starter工程hello-spring-boot-starter并配置pom.xml文件 <?xml version"1.0" encoding"UTF-8"?> <project xmlns"http://maven.apache.org/POM/4.0.0"xmlns:xsi"http://www.w3.org/2001/XMLSchem…

SpringBoot优点达项目实战:项目初始化(一)

SpringBoot优点达项目实战&#xff1a;项目初始化&#xff08;一&#xff09; 文章目录 SpringBoot优点达项目实战&#xff1a;项目初始化&#xff08;一&#xff09;1、项目介绍2、项目搭建3、依赖导入4、数据准备 1、项目介绍 技术框架 SpringbootmybatisPlusvueknife 2、项目…

什么是生成式AI?

生成式AI&#xff08;Generative AI&#xff09;是一类利用机器学习和人工智能技术来生成内容的系统。这些系统可以创建文本、图像、音乐、视频等各种类型的内容。生成式AI通过学习大量的数据来理解和模仿人类的创作过程&#xff0c;从而生成新的、原创的内容。以下是生成式AI的…

创建App

自学python如何成为大佬(目录):https://blog.csdn.net/weixin_67859959/article/details/139049996?spm1001.2014.3001.5501 在Django项目中&#xff0c;推荐使用App来完成不同模块的任务&#xff0c;通过执行如下命令可以启用一个应用程序。 python manage.py startapp app…