Fusing Global and Local Features for Generalized AI-Synthesized Image Detection
Not on the map. Why?
Abstract (arXiv)
With the development of the Generative Adversarial Networks (GANs) and DeepFakes, AI-synthesized images are now of such high quality that humans can hardly distinguish them from real images. It is imperative for media forensics to develop detectors to expose them accurately. Existing detection methods have shown high performance in generated images detection, but they tend to generalize poorly in the real-world scenarios, where the synthetic images are usually generated with unseen models using unknown source data. In this work, we emphasize the importance of combining information from the whole image and informative patches in improving the generalization ability of AI-synthesized image detection. Specifically, we design a two-branch model to combine global spatial information from the whole image and local informative features from multiple patches selected by a novel patch selection module. Multi-head attention mechanism is further utilized to fuse the global and local features. We collect a highly diverse dataset synthesized by 19 models with various objects and resolutions to evaluate our model. Experimental results demonstrate the high accuracy and good generalization ability of our method in detecting generated images. Our code is available at https://github.com/littlejuyan/FusingGlobalandLocal.
Key facts. Cited by 88 papers in the map (data through 14 Oct 2026). Builds on 35 of its 41 references that are in the map. First posted on arXiv 26 Mar 2022.
Builds on
35 of 41 references.
References · 35
- 2017Attention Is All You Needcited by 60,361
- 2015Deep Residual Learning for Image Recognitioncited by 52,096
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
- 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
- 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,527
- 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
- 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,253
- 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,988
- 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
- 2019Semantic Image Synthesis with Spatially-Adaptive Normalizationcited by 1,299
- 2021Alias-Free Generative Adversarial Networkscited by 1,184
- 2019FaceForensics++: Learning to Detect Manipulated Facial Imagescited by 1,101
23 more
- 2017StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translationcited by 1,039
- 2019CNN-generated images are surprisingly easy to spot... for nowcited by 847
- 2019Celeb-DF: A Large-scale Challenging Dataset for DeepFake Forensicscited by 575
- 2018Learning to See in the Darkcited by 460
- 2015Autoencoding beyond pixels using a learned similarity metriccited by 407
- 2021Multi-attentional Deepfake Detectioncited by 341
- 2020What makes fake images detectable? Understanding properties that generalizecited by 273
- 2020The Creation and Detection of Deepfakes: A Surveycited by 247
- 2017Photographic Image Synthesis with Cascaded Refinement Networkscited by 223
- 2021Are GAN generated images easy to detect? A critical analysis of the state-of-the-artcited by 139
- 2020MODNet: Real-Time Trimap-Free Portrait Matting via Objective Decompositioncited by 127
- 2018Diverse Image Synthesis from Semantic Layouts via Conditional IMLEcited by 60
- 2019Incremental learning for the detection and classification of GAN-generated imagescited by 58
- 2021Reverse Engineering of Generative Models: Inferring Model Hyperparameters from Generated Imagescited by 58
- 2020DeepFake Detection: Current Challenges and Next Stepscited by 44
- 2021FReTAL: Generalizing Deepfake Detection using Knowledge Distillation and Representation Learningcited by 44
- 2020Detecting CNN-Generated Facial Images in Real-World Scenarioscited by 32
- 2022GAN-generated Faces Detection: A Survey and New Perspectivescited by 30
- 2020T-GD: Transferable GAN-generated Images Detection Frameworkcited by 28
- 2021Robust Attentive Deep Neural Network for Exposing GAN-generated Facescited by 28
- 2021Towards Universal GAN Image Detectioncited by 26
- 2020Deepfake detection: humans vs. machinescited by 14
- 2021Finding Facial Forgery Artifacts with Parts-Based Detectorscited by 14
Built on by
Citing papers per year: 2022: 2; 2023: 6; 2024: 21; 2025: 31; 2026: 28.
88 papers.
Sort:
By area:
- 2022On the detection of synthetic images generated by diffusion modelscited by 283
- 2024A Sanity Check for AI-generated Image Detectioncited by 189
- 2023Raising the Bar of AI-generated Image Detection with CLIPcited by 164
- 2024VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Modelscited by 131
- 2023PatchCraft: Exploring Texture Patch for Efficient AI-generated Image Detectioncited by 122
- 2024SIDA: Social Media Image Deepfake Detection, Localization and Explanation with Large Multimodal Modelcited by 108
- 2024Improving Synthetic Image Detection Towards Generalization: An Image Transformation Perspectivecited by 89
- 2024Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensicscited by 69
- 2025Spot the Fake: Large Multimodal Model-Based Synthetic Image Detection with Artifact Explanationcited by 68
- 2024FakeInversion: Learning to Detect Images from Unseen Text-to-Image Models by Inverting Stable Diffusioncited by 61
- 2024Any-Resolution AI-Generated Image Detection by Spectral Learningcited by 57
- 2023Generalizable Synthetic Image Detection via Language-guided Contrastive Learningcited by 49
38 more
- 2023AutoSplice: A Text-prompt Manipulated Image Dataset for Media Forensicscited by 46
- 2025CO-SPY: Combining Semantic and Pixel Features to Detect Synthetic Images by AIcited by 42
- 2024FakeBench: Probing Explainable Fake Image Detection via Large Multimodal Modelscited by 36
- 2025AIGI-Holmes: Towards Explainable and Generalizable AI-Generated Image Detection via Multimodal Large Language Modelscited by 33
- 2022GLFF: Global and Local Feature Fusion for AI-synthesized Image Detectioncited by 25
- 2025Bridging the Gap Between Ideal and Real-world Evaluation: Benchmarking AI-Generated Image Detection in Challenging Scenarioscited by 24
- 2024TGIF: Text-Guided Inpainting Forgery Datasetcited by 20
- 2023Improving Synthetically Generated Image Detection in Cross-Concept Settingscited by 15
- 2026Your One-Stop Solution for AI-Generated Video Detectioncited by 15
- 2024Synthetic Image Verification in the Era of Generative AI: What Works and What Isn't There Yetcited by 13
- 2024Uncertainty-Guided Appearance-Motion Association Network for Out-of-Distribution Action Detectioncited by 12
- 2025Methods and Trends in Detecting AI-Generated Images: A Comprehensive Reviewcited by 12
- 2024Data-Independent Operator: A Training-Free Artifact Representation Extractor for Generalizable Deepfake Detectioncited by 11
- 2025Navigating the Challenges of AI-Generated Image Detection in the Wild: What Truly Matters?cited by 10
- 2024TextureCrop: Enhancing Synthetic Image Detection through Texture-based Croppingcited by 9
- 2024TIP-I2V: A Million-Scale Real Text and Image Prompt Dataset for Image-to-Video Generationcited by 9
- 2024Evolution of Detection Performance throughout the Online Lifespan of Synthetic Imagescited by 8
- 2025Rethinking Cross-Generator Image Forgery Detection through DINOv3cited by 8
- 2024Detecting Origin Attribution for Text-to-Image Diffusion Modelscited by 7
- 2025DFBench: Benchmarking Deepfake Image Detection Capability of Large Multimodal Modelscited by 7
- 2026Simplicity Prevails: The Emergence of Generalizable AIGI Detection in Visual Foundation Modelscited by 7
- 2024Continuous fake media detection: adapting deepfake detectors to new generative techniquescited by 6
- 2024ODDN: Addressing Unpaired Data Challenges in Open-World Deepfake Detection on Online Social Networkscited by 6
- 2024FairAdapter: Detecting AI-generated Images with Improved Fairnesscited by 6
- 2025Self-Supervised AI-Generated Image Detection: A Camera Metadata Perspectivecited by 6
- 2025CausalCLIP: Causally-Informed Feature Disentanglement and Filtering for Generalizable Detection of Generated Imagescited by 6
- 2023Synthetic Image Detection: Highlights from the IEEE Video and Image Processing Cup 2022 Student Competitioncited by 5
- 2024Artifact Feature Purification for Cross-domain Detection of AI-generated Imagescited by 5
- 2025Multimodal Conditional Information Bottleneck for Generalizable AI-Generated Image Detectioncited by 5
- 2025Pay Less Attention to Deceptive Artifacts: Robust Detection of Compressed Deepfakes on Online Social Networkscited by 5
- 2025DINO-Detect: A Simple yet Effective Framework for Blur-Robust AI-Generated Image Detectioncited by 5
- 2026Where Detectors Fail: Probing Generative Space for Generalizable AI-Generated Image Detectioncited by 5
- 2025SFLD: Reducing the content bias for AI-generated Image Detectioncited by 4
- 2026FakeVLM-R1: Internalizing Physical Laws via CoT for Synthetic Image Detectioncited by 4
- 2024DeepFake-O-Meter v2.0: An Open Platform for DeepFake Detectioncited by 3
- 2025Can GPT tell us why these images are synthesized? Empowering Multimodal Large Language Models for Forensicscited by 3
- 2025When Semantics Regulate: Rethinking Patch Shuffle and Internal Bias for Generated Image Detection with CLIPcited by 3
- 2026VIGIL: Part-Grounded Structured Reasoning for Generalizable Deepfake Detectioncited by 3
50 most cited of 88 All 88 (JSON)
- 2026Trustworthy synthetic visual media: Evidence across the media lifecycle28 Sep 2026
- 2026Unifying Semantic Priors and High-Frequency Traces: Enhancing V-JEPA with Mixture-of-Experts for Robust Synthetic Image Forensics15 Sep 2026
- 2026From Detection to Localization: A Unified Forensics Framework for Fully Synthetic and Tampered Images2 Sep 2026
- 2026UC-VLM: Consistency-Driven Learning for AI-Generated Image Detection with Vision-Language Large Models15 Aug 2026
- 2026Generated Images Are Easier to Forget: A Machine Unlearning Perspective for Synthetic Image Detection1 Aug 2026
- 2026FAIR: Feature-Augmented Implicit Regularization for AI-generated Fake Image Detection24 Jul 2026
- 2026Perception, Verdict, and Evolution: Hindsight-Driven Self-Refining Forensics Agent for AI-Generated Image Detection25 Jun 2026
- 2026Minimalist Preprocessing Approach for Image Synthesis Detection24 Jun 2026
- 2026ForensicConcept: Transferable Forensic Concepts for AIGI Detection5 Jun 2026
- 2026FakeVLM-R1: Internalizing Physical Laws via CoT for Synthetic Image Detection28 May 2026
- 2026HydraPrompt: An Adaptive and Asymmetric Framework of Vision-Language Models for Synthetic Image Detection26 May 2026
- 2026Where Detectors Fail: Probing Generative Space for Generalizable AI-Generated Image Detection24 May 2026
38 more
- 2026GenShield: Unified Detection and Artifact Correction for AI-Generated Images15 May 2026
- 2026LEGO: LoRA-Enabled Generator-Oriented Framework for Synthetic Image Detection6 May 2026
- 2026HiMix: Hierarchical Artifact-aware Mixup for Generalized Synthetic Image Detection30 Apr 2026
- 2026UniGenDet: A Unified Generative-Discriminative Framework for Co-Evolutionary Image Generation and Generated Image Detection23 Apr 2026
- 2026Combating Pattern and Content Bias: Adversarial Feature Learning for Generalized AI-Generated Image Detection14 Apr 2026
- 2026Detecting Diffusion-generated Images via Dynamic Assembly Forests10 Apr 2026
- 2026VIGIL: Part-Grounded Structured Reasoning for Generalizable Deepfake Detection23 Mar 2026
- 2026From Masks to Pixels and Meaning: A New Taxonomy, Benchmark, and Metrics for VLM Image Tampering20 Mar 2026
- 2026SimLBR: Learning to Detect Fake Images by Learning to Detect Real Images23 Feb 2026
- 2026FOCA: Frequency-Oriented Cross-Domain Forgery Detection, Localization and Explanation via Multi-Modal Large Language Model21 Feb 2026
- 2026Your AI-Generated Image Detector Can Secretly Achieve SOTA Accuracy, If Calibrated2 Feb 2026
- 2026Simplicity Prevails: The Emergence of Generalizable AIGI Detection in Visual Foundation Models2 Feb 2026
- 2026MPF-Net: Exposing High-Fidelity AI-Generated Video Forgeries via Hierarchical Manifold Deviation and Micro-Temporal Fluctuations29 Jan 2026
- 2026S^2F-Net:A Robust Spatial-Spectral Fusion Framework for Cross-Model AIGC Detection18 Jan 2026
- 2026Your One-Stop Solution for AI-Generated Video Detection16 Jan 2026
- 2026Attention to Detail: Global-Local Attention for High-Resolution AI-Generated Image Detection1 Jan 2026
- 2025Patch-Discontinuity Mining for Generalized Deepfake Detection26 Dec 2025
- 2025Detection of AI Generated Images Using Combined Uncertainty Measures and Particle Swarm Optimised Rejection Mechanism20 Dec 2025
- 2025Detecting Localized Deepfakes: How Well Do Synthetic Image Detectors Handle Inpainting?18 Dec 2025
- 2025CausalCLIP: Causally-Informed Feature Disentanglement and Filtering for Generalizable Detection of Generated Images15 Dec 2025
- 2025Self-Supervised AI-Generated Image Detection: A Camera Metadata Perspective5 Dec 2025
- 2025Rethinking Cross-Generator Image Forgery Detection through DINOv327 Nov 2025
- 2025AVFakeBench: A Comprehensive Audio-Video Forgery Detection Benchmark for AV-LMMs26 Nov 2025
- 2025When Semantics Regulate: Rethinking Patch Shuffle and Internal Bias for Generated Image Detection with CLIP24 Nov 2025
- 2025Beyond Binary Classification: A Semi-supervised Approach to Generalized AI-generated Image Detection23 Nov 2025
- 2025DINO-Detect: A Simple yet Effective Framework for Blur-Robust AI-Generated Image Detection16 Nov 2025
- 2025Detecting AI-Generated Images via Contextual Anomaly Estimation in Masked AutoEncoders9 Nov 2025
- 2025Mirage: Unveiling Hidden Artifacts in Synthetic Images with Large Vision-Language Models4 Oct 2025
- 2025DF-LLaVA: Unlocking MLLMs for Synthetic Image Detection via Knowledge Injection and Conflict-Driven Self-Reflection18 Sep 2025
- 2025Bridging the Gap Between Ideal and Real-world Evaluation: Benchmarking AI-Generated Image Detection in Challenging Scenarios11 Sep 2025
- 2025A Novel Local Focusing Mechanism for Deepfake Detection Generalization23 Aug 2025
- 2025MiraGe: Multimodal Discriminative Representation Learning for Generalizable AI-Generated Image Detection3 Aug 2025
- 2025Navigating the Challenges of AI-Generated Image Detection in the Wild: What Truly Matters?14 Jul 2025
- 2025AIGI-Holmes: Towards Explainable and Generalizable AI-Generated Image Detection via Multimodal Large Language Models3 Jul 2025
- 2025Pay Less Attention to Deceptive Artifacts: Robust Detection of Compressed Deepfakes on Online Social Networks25 Jun 2025
- 2025DFBench: Benchmarking Deepfake Image Detection Capability of Large Multimodal Models3 Jun 2025
- 2025Multimodal Conditional Information Bottleneck for Generalizable AI-Generated Image Detection21 May 2025
- 2025Can GPT tell us why these images are synthesized? Empowering Multimodal Large Language Models for Forensics16 Apr 2025
50 newest of 88 All 88 (JSON)
- 2022On the detection of synthetic images generated by diffusion modelscited by 283
- 2024VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Modelscited by 131
- 2024FakeInversion: Learning to Detect Images from Unseen Text-to-Image Models by Inverting Stable Diffusioncited by 61
- 2023Generalizable Synthetic Image Detection via Language-guided Contrastive Learningcited by 49
- 2022GLFF: Global and Local Feature Fusion for AI-synthesized Image Detectioncited by 25
- 2025Methods and Trends in Detecting AI-Generated Images: A Comprehensive Reviewcited by 12
- 2024Detecting Origin Attribution for Text-to-Image Diffusion Modelscited by 7
- 2024Artifact Feature Purification for Cross-domain Detection of AI-generated Imagescited by 5
- 2025Pay Less Attention to Deceptive Artifacts: Robust Detection of Compressed Deepfakes on Online Social Networkscited by 5
- 2025When Semantics Regulate: Rethinking Patch Shuffle and Internal Bias for Generated Image Detection with CLIPcited by 3
- 2026GenShield: Unified Detection and Artifact Correction for AI-Generated Imagescited by 1
- 2026Detecting Diffusion-generated Images via Dynamic Assembly Forestscited by 0
- 2026GenShield: Unified Detection and Artifact Correction for AI-Generated Images15 May 2026
- 2026Detecting Diffusion-generated Images via Dynamic Assembly Forests10 Apr 2026
- 2025When Semantics Regulate: Rethinking Patch Shuffle and Internal Bias for Generated Image Detection with CLIP24 Nov 2025
- 2025Pay Less Attention to Deceptive Artifacts: Robust Detection of Compressed Deepfakes on Online Social Networks25 Jun 2025
- 2025Methods and Trends in Detecting AI-Generated Images: A Comprehensive Review21 Feb 2025
- 2024FakeInversion: Learning to Detect Images from Unseen Text-to-Image Models by Inverting Stable Diffusion12 Jun 2024
- 2024Detecting Origin Attribution for Text-to-Image Diffusion Models28 Mar 2024
- 2024Artifact Feature Purification for Cross-domain Detection of AI-generated Images17 Mar 2024
- 2024VidProM: A Million-scale Real Prompt-Gallery Dataset for Text-to-Video Diffusion Models10 Mar 2024
- 2023Generalizable Synthetic Image Detection via Language-guided Contrastive Learning23 May 2023
- 2022GLFF: Global and Local Feature Fusion for AI-synthesized Image Detection16 Nov 2022
- 2022On the detection of synthetic images generated by diffusion models1 Nov 2022
- 2023Raising the Bar of AI-generated Image Detection with CLIPcited by 164
- 2024SIDA: Social Media Image Deepfake Detection, Localization and Explanation with Large Multimodal Modelcited by 108
- 2024Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensicscited by 69
- 2025Spot the Fake: Large Multimodal Model-Based Synthetic Image Detection with Artifact Explanationcited by 68
- 2024FakeBench: Probing Explainable Fake Image Detection via Large Multimodal Modelscited by 36
- 2025AIGI-Holmes: Towards Explainable and Generalizable AI-Generated Image Detection via Multimodal Large Language Modelscited by 33
- 2025DFBench: Benchmarking Deepfake Image Detection Capability of Large Multimodal Modelscited by 7
- 2025Can GPT tell us why these images are synthesized? Empowering Multimodal Large Language Models for Forensicscited by 3
- 2025Mirage: Unveiling Hidden Artifacts in Synthetic Images with Large Vision-Language Modelscited by 1
- 2026HydraPrompt: An Adaptive and Asymmetric Framework of Vision-Language Models for Synthetic Image Detectioncited by 1
- 2026HydraPrompt: An Adaptive and Asymmetric Framework of Vision-Language Models for Synthetic Image Detection26 May 2026
- 2025Mirage: Unveiling Hidden Artifacts in Synthetic Images with Large Vision-Language Models4 Oct 2025
- 2025AIGI-Holmes: Towards Explainable and Generalizable AI-Generated Image Detection via Multimodal Large Language Models3 Jul 2025
- 2025DFBench: Benchmarking Deepfake Image Detection Capability of Large Multimodal Models3 Jun 2025
- 2025Can GPT tell us why these images are synthesized? Empowering Multimodal Large Language Models for Forensics16 Apr 2025
- 2025Spot the Fake: Large Multimodal Model-Based Synthetic Image Detection with Artifact Explanation19 Mar 2025
- 2024SIDA: Social Media Image Deepfake Detection, Localization and Explanation with Large Multimodal Model5 Dec 2024
- 2024FakeBench: Probing Explainable Fake Image Detection via Large Multimodal Models20 Apr 2024
- 2024Can ChatGPT Detect DeepFakes? A Study of Using Multimodal Large Language Models for Media Forensics21 Mar 2024
- 2023Raising the Bar of AI-generated Image Detection with CLIP30 Nov 2023
- 2024A Sanity Check for AI-generated Image Detectioncited by 189
- 2023PatchCraft: Exploring Texture Patch for Efficient AI-generated Image Detectioncited by 122
- 2024Improving Synthetic Image Detection Towards Generalization: An Image Transformation Perspectivecited by 89
- 2024Any-Resolution AI-Generated Image Detection by Spectral Learningcited by 57
- 2023AutoSplice: A Text-prompt Manipulated Image Dataset for Media Forensicscited by 46
- 2025CO-SPY: Combining Semantic and Pixel Features to Detect Synthetic Images by AIcited by 42
- 2025Bridging the Gap Between Ideal and Real-world Evaluation: Benchmarking AI-Generated Image Detection in Challenging Scenarioscited by 24
- 2024TGIF: Text-Guided Inpainting Forgery Datasetcited by 20
- 2023Improving Synthetically Generated Image Detection in Cross-Concept Settingscited by 15
- 2026Your One-Stop Solution for AI-Generated Video Detectioncited by 15
- 2024Synthetic Image Verification in the Era of Generative AI: What Works and What Isn't There Yetcited by 13
- 2024Uncertainty-Guided Appearance-Motion Association Network for Out-of-Distribution Action Detectioncited by 12
12 most cited of 61 All 61 (JSON)
- 2026Trustworthy synthetic visual media: Evidence across the media lifecycle28 Sep 2026
- 2026From Detection to Localization: A Unified Forensics Framework for Fully Synthetic and Tampered Images2 Sep 2026
- 2026UC-VLM: Consistency-Driven Learning for AI-Generated Image Detection with Vision-Language Large Models15 Aug 2026
- 2026FAIR: Feature-Augmented Implicit Regularization for AI-generated Fake Image Detection24 Jul 2026
- 2026Minimalist Preprocessing Approach for Image Synthesis Detection24 Jun 2026
- 2026ForensicConcept: Transferable Forensic Concepts for AIGI Detection5 Jun 2026
- 2026FakeVLM-R1: Internalizing Physical Laws via CoT for Synthetic Image Detection28 May 2026
- 2026Where Detectors Fail: Probing Generative Space for Generalizable AI-Generated Image Detection24 May 2026
- 2026LEGO: LoRA-Enabled Generator-Oriented Framework for Synthetic Image Detection6 May 2026
- 2026HiMix: Hierarchical Artifact-aware Mixup for Generalized Synthetic Image Detection30 Apr 2026
- 2026UniGenDet: A Unified Generative-Discriminative Framework for Co-Evolutionary Image Generation and Generated Image Detection23 Apr 2026
- 2026Combating Pattern and Content Bias: Adversarial Feature Learning for Generalized AI-Generated Image Detection14 Apr 2026
12 newest of 61 All 61 (JSON)