Grounding and Explaining Visual Evidence for AI-Generated Image Detection in Human-Centric Scenes
2026-08-03 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors created a large dataset called HAVE with real and AI-generated images focused on people, including detailed notes on which parts of the images show evidence of being fake. They also developed a new method, PAVE, that can tell if an image is real or AI-made, highlight the fake parts, and explain why those parts look suspicious. Their approach checks if the explanations really match the highlighted areas and makes sure the model pays attention to image details by comparing normal and masked images. Tests show their method works well in spotting fakes and providing clear reasons. They will share their dataset and code after publishing.
AI-generated imagesAuthenticity detectionVisual evidence groundingExplainable AIHuman-centric scenesBounding box annotationRegion-aligned explanationPerception-aware regularizationDatasetImage generation models
Authors
Kun Guo, Yuzhou Yang, Haoyue Wang, Qichao Ying, Sheng Li, Zhenxing Qian
Abstract
Rapid advances in image generation models call for interpretable AI-generated image detection methods that not only determine authenticity but also provide supporting visual evidence. Existing approaches may produce inconsistencies between generated explanations and localized evidence regions, undermining the reliability of explanations for authenticity decisions. Meanwhile, existing benchmarks provide limited coverage of the diverse human-centric scenes prevalent in generated imagery. To address these limitations, we investigate authenticity detection with grounded and explainable visual evidence in human-centric scenes. We present HAVE (Human-centric AI-generated Visual Evidence), a diverse human-centric dataset comprising 40K real and 39K AI-generated images from 10 recent generators, with 106K localized evidence instances across 8 evidence categories, each annotated with a bounding box and a region-aligned explanation. We further propose PAVE, a Perception-Aware Visual Evidence framework that jointly performs authenticity prediction, visual evidence grounding, and region-aligned explanation generation. PAVE employs a judge-guided alignment reward to assess region--explanation consistency and evidence validity, together with perception-aware regularization that contrasts token-level predictions between original and randomly masked images to promote reliance on visual input. Experiments on HAVE and external datasets demonstrate strong performance in authenticity detection, visual evidence grounding, and explanation quality. Code and data will be released upon publication.