本文取自笔者参与合作的github repo文档,并分工于大部分根据论文列表查阅搜集相关材料的工作。该repo的仓库地址为🔗;该repo基于一已有项目戳这里构建(给前辈们撒花致谢🎉),并且对截止2025.7.2之前的AIGC检测相关论文进行整理。
照片本是记录生活的介质,如今却因AI技术面临”信任危机”——AIGC的飞速发展,让”PS都弱爆了”的时代真正来临。当眼见不再为实,如何鉴别AI生成的”照骗”已成为AI安全领域的头号挑战之一。本清单系统整理AIGC图像检测相关研究,希望对大家有所帮助。
📚Datasets
Year | Dataset | Number of Real | Number of Fake | Source of Real Image | Generation Method of Fake Image |
---|---|---|---|---|---|
2020 | CNNSpot | 362,000 | 362,000 | LSUN, ImageNet, CelebA, COCO… | ProGAN, StyleGAN, BigGAN, CRN, SITD… |
2023 | GenImage | 1,331,167 | 1,350,000 | ImagNet | SDMs, Midjourney, BigGAN |
2023 | Fake2M | - | 2,300,000 | CC3M | SD-V1.5, IF, StyleGAN3 |
2023 | DMimage | 200,000 | 200,000 | COOC, LSUN | LDM |
2023 | DiffusionDB | 3,300,000 | 16,000,000 | DiscordChatExporter | SD |
2024 | WildFake | 2,557,278 | 1,013,446 | ImagNet, Laion, Wukong, COO… | BigGAN, StyleGAN, StarGAN, Midjourney, DALLE… |
📝Papers
2️⃣0️⃣2️⃣5️⃣
Few-Shot Learner Generalizes Across AI-Generated Image Detection
[Paper] [Code]Forensic Self-Descriptions Are All You Need for Zero-Shot Detection, Open-Set Source Attribution, and Clustering of AI-generated Images
[Paper]Beyond Generation: A Diffusion-based Low-level Feature Extractor for Detecting AI-generated Images
[Paper]Secret Lies in Color: Enhancing AI-Generated Images Detection with Color Distribution Analysis
[Paper]Towards Universal AI-Generated Image Detection by Variational Information Bottleneck Network
[Paper]Where’s the Liability in the Generative Era? Recovery-based Black-Box Detection of AI-Generated Content
[Paper]FIRE: Robust Detection of Diffusion-Generated Images via Frequency-Guided Reconstruction Error
[Paper]Forensic Self-Descriptions Are All You Need for Zero-Shot Detection, Open-Set Source Attribution, and Clustering of AI-generated Images
[Paper]Orthogonal Subspace Decomposition for Generalizable AI-Generated Image Detection
[Paper] [code]D3 Scaling Up Deepfake Detection by Learning
[Paper] [Code]SIDA: Social Media Image Deepfake Detection, Localization and Explanation with Large Multimodal Model
[Paper] [Code]A Bias-Free Training Paradigm for More General AI-generated Image Detection
[Paper] [Code]Any-Resolution AI-Generated Image Detection by Spectral Learning
[Paper] [Code]TextureCrop: Enhancing Synthetic Image Detection through Texture-based Cropping
[Paper] [Code]HEIE: MLLM-Based Hierarchical Explainable AIGC Image Implausibility Evaluator
[Paper] [Datasets] [Code]Improving Synthetic Image Detection Towards Generalization: An Image Transformation Perspective
[Paper] [Code]SFLD: Reducing the content bias for AI-generated Image Detection
[Paper]A SANITY CHECK FOR AI-GENERATED IMAGE DETECTION
[Paper] [Code]FAKESHIELD: EXPLAINABLE IMAGE FORGERY DETECTION AND LOCALIZATION VIA MULTI-MODAL LARGE LANGUAGE MODELS
[Paper] [Code]Exploring the Collaborative Advantage of Low-level Information on Generalizable AI-Generated Image Detection
[Paper]HYPERDET: GENERALIZABLE DETECTION OF SYNTHESIZED IMAGES BY GENERATING AND MERGING A MIXTURE OF HYPER LORAS
[Paper]Zooming In on Fakes: A Novel Dataset for Localized AI-Generated Image Detection with Forgery Amplification Approach
[Paper] [Code]AntifakePrompt: Prompt-Tuned Vision-Language Models are Fake Image Detectors
[Paper] [Code]Can GPT tell us why these images are synthesized? Empowering Multimodal Large Language Models for Forensics
[Paper]Exploring Modality Disruption in Multimodal Fake News Detection
[Paper]
2️⃣0️⃣2️⃣4️⃣
FakeInversion: Learning to Detect Images from Unseen Text-to-Image Models by Inverting Stable Diffusion
[Paper] [Code]Forgery-aware Adaptive Transformer for Generalizable Synthetic
[Paper] [Code]Rethinking the Up-Sampling Operations in CNN-based Generative Network for Generalizable Deepfake Detection
[Paper] [Code]LaRE2: Latent Reconstruction Error Based Method for Diffusion-Generated Image Detection
[Paper] [Code]AEROBLADE: Training-Free Detection of Latent Diffusion Images Using Autoencoder Reconstruction Error
[Paper] [Code]MiRAGeNews: Multimodal Realistic AI-Generated News Detection
[Paper] [Code]Zero-Shot Detection of AI-Generated Images
[Paper] [Code]Leveraging Representations from Intermediate Encoder-blocks for Synthetic Image Detection
[Paper] [Code]Contrasting Deepfakes Diffusion via Contrastive Learning and Global-Local Similarities
[Paper] [Code]DRCT: Diffusion Reconstruction Contrastive Training towards Universal Detection of Diffusion Generated Images [Paper] [Code]
Exposing the Fake: Effective Diffusion-Generated Images Detection
[Paper] [Code]How to Trace Latent Generative Model Generated Images without Artificial Watermark?
[Paper] [Code]CLIPping the Deception: Adapting Vision-Language Models for Universal Deepfake Detection
[Paper] [Code]Harnessing the Power of Large Vision Language Models for Synthetic Image Detection
[Paper] [Code]DE-FAKE: Detection and Attribution of Fake Images Generated by Text-to-Image Generation Models
[Paper] [Code]Raising the Bar of AI-generated Image Detection with CLIP
[Paper] [Code]Faster Than Lies: Real-time Deepfake Detection using Binary Neural Networks
[Paper] [Code]Breaking Semantic Artifacts for Generalized AI-generated Image Detection
[Paper] [Code]Mixture of Low-rank Experts for Transferable AI-Generated Image Detection
[Paper] [Code]Guided and Fused: Efficient Frozen CLIP-ViT with Feature Guidance and Multi-Stage Feature Fusion for Generalizable Deepfake Detection
[Paper]A Single Simple Patch is All You Need for AI-generated Image Detection
[Paper] [Code]RIGID: A Training-Free and Model-Agnostic Framework for Robust AI-Generated Image Detection
[Paper]Improving Interpretability and Robustness for the Detection of AI-Generated Images
[Paper]Continuous fake media detection: adapting deepfake detectors to new generative techniques
[Paper]Organic or Diffused: Can We Distinguish Human Art from AI-generated Images?
[Paper]Let Real Images be as a Judger, Spotting Fake Images Synthesized with Generative Models
[Paper]Detecting Image Attribution for Text-to-Image Diffusion Models in RGB and Beyond
[Paper] [Code] [Code]MLEP: Multi-granularity Local Entropy Patterns for Generalized AI-generated Image Detection
[Paper]FFAA: Multimodal Large Language Model based Explainable Open-World Face Forgery Analysis Assistant
[Paper] [Code]
2️⃣0️⃣2️⃣3️⃣
Towards Universal Fake Image Detectors that Generalize Across Generative Models
[Paper] [Code]Learning on Gradients: Generalized Artifacts Representation for GAN-Generated Images Detection
[Paper] [Code]DIRE for Diffusion-Generated Image Detection
[Paper] [Code]Seeing is not always believing: Benchmarking Human and Model Perception of AI-Generated Images
[Paper] [Code]On The Detection of Synthetic Images Generated by Diffusion Models
[Paper] [Code]DE-FAKE: Detection and Attribution of Fake Images Generated by Text-to-Image Generation Models
[Paper]Synthbuster: Towards Detection of Diffusion Model Generated Images
[Paper]Online Detection of AI-Generated Images
[Paper]Detecting Images Generated by Deep Diffusion Models using their Local Intrinsic Dimensionality
[Paper]PatchCraft: Exploring Texture Patch for Efficient AI-generated Image Detection
[Paper]Generalizable Synthetic Image Detection via Language-guided Contrastive Learning
[Paper] [Code]Raising the Bar of AI-generated Image Detection with CLIP
[Paper] [Code]GenDet: Towards Good Generalizations for AI-Generated Image Detection
[Paper]AntifakePrompt: Prompt-Tuned Vision-Language Models are Fake Image Detectors
[Paper]Generalizable Synthetic Image Detection via Language-guided Contrastive Learning
[Paper] [Code]PatchCraft: Exploring Texture Patch for Efficient AI-generated Image Detection
[Paper] [Code]
2️⃣0️⃣2️⃣2️⃣
Detecting Generated Images by Real Images
[Paper] [Code]FingerprintNet: Synthesized Fingerprints for Generated Image Detection
[Paper]
2️⃣0️⃣2️⃣1️⃣
Are GAN generated images easy to detect? A critical analysis of the state-of-the-art
[Paper]
2️⃣0️⃣2️⃣0️⃣
CNN-generated images are surprisingly easy to spot… for now
[Paper] [Code]Global Texture Enhancement for Fake Face Detection In the Wild
[Paper] [Code]Watch your Up-Convolution: CNN Based Generative Deep Neural Networks are Failing to Reproduce Spectral Distributions
[Paper] [Code]What makes fake images detectable? Understanding properties that generalize
[Paper] [Code]FingerprintNet: Synthesized Fingerprints for Generated Image Detection
[Paper] [Code]🏷️(代码貌似非官方)