SOTAVerified

Image Manipulation Detection

The task of detecting images or image parts that have been tampered or manipulated (sometimes also referred to as doctored). This typically encompasses image splicing, copy-move, or image inpainting.

Papers

Showing 110 of 73 papers

TitleStatusHype
Weakly-supervised Localization of Manipulated Image Regions Using Multi-resolution Learned Features0
ForensicHub: A Unified Benchmark & Codebase for All-Domain Fake Image Detection and LocalizationCode2
AnimeDL-2M: Million-Scale AI-Generated Anime Image Detection and Localization in Diffusion Era0
Context-Aware Weakly Supervised Image Manipulation Localization with SAM Refinement0
LEGION: Learning to Ground and Explain for Synthetic Image Detection0
IMDPrompter: Adapting SAM to Image Manipulation Detection by Cross-View Automated Prompt Learning0
Data-Driven Fairness Generalization for Deepfake Detection0
ForgerySleuth: Empowering Multimodal Large Language Models for Image Manipulation DetectionCode1
HRGR: Enhancing Image Manipulation Detection via Hierarchical Region-aware Graph Reasoning0
Perturb, Attend, Detect and Localize (PADL): Robust Proactive Image Defense0
Show:102550
← PrevPage 1 of 8Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1TruForBalanced Accuracy0.98Unverified
2Early FusionBalanced Accuracy0.96Unverified
3DF-NetAUC0.88Unverified
4Late FusionBalanced Accuracy0.82Unverified
5CAT-Net v2Balanced Accuracy0.8Unverified
6MVSS-NetBalanced Accuracy0.73Unverified
7CR-CNNBalanced Accuracy0.63Unverified
8ManTraNetBalanced Accuracy0.5Unverified