deepfakes
NTIRE-RobustAIGenDetection-train
Training set for NTIRE 2026 Robust AI-Generated Image Detection in the Wild
Robust AI-Generated Image Detection in the Wild Challenge is organized as a part of the New Trends in Image Restoration and Enhancement Workshop in conjunction with CVPR 2026.
Challenge overview
Text-to-image (T2I) models have made synthetic images nearly indistinguishable from real photos in many cases, which creates serious challenges for trust, authenticity, forensics, and content… See the full description on the dataset page: https://huggingface.co/datasets/deepfakesMSU/NTIRE-RobustAIGenDetection-train.CVQAD
MSU Compression Dataset Description
We developed LEHA-CVQAD dataset to evaluate full-reference and no-reference video quality metrics. Here we share the open part of the whole compression artifacts dataset (1,962 out of 6,240 videos). The hidden part is only available to benchmark-support personnel for testing metric performance. All videos are of mostly FullHD resolution, YUV420, and 10-15 seconds duration. Fps values are 24, 25, 30, 39, 50, and 60.
Subjective quality scores are… See the full description on the dataset page: https://huggingface.co/datasets/deepfakesMSU/CVQAD.popular-deepfakesNTIRE-RobustAIGenDetection-val
Validation set for NTIRE 2026 Robust AI-Generated Image Detection in the Wild (updated)
Note: This is an updated version of the dataset. For challenge submissions, please make sure you use this version.
Robust AI-Generated Image Detection in the Wild Challenge is organized as a part of the New Trends in Image Restoration and Enhancement Workshop in conjunction with CVPR 2026.
Challenge overview
Text-to-image (T2I) models have made synthetic images nearly… See the full description on the dataset page: https://huggingface.co/datasets/deepfakesMSU/NTIRE-RobustAIGenDetection-val.NTIRE-RobustAIGenDetection-test-public
Test set for NTIRE 2026 Robust AI-Generated Image Detection in the Wild
Robust AI-Generated Image Detection in the Wild Challenge is organized as a part of the New Trends in Image Restoration and Enhancement Workshop in conjunction with CVPR 2026.
Challenge overview
Text-to-image (T2I) models have made synthetic images nearly indistinguishable from real photos in many cases, which creates serious challenges for trust, authenticity, forensics, and content safety. At… See the full description on the dataset page: https://huggingface.co/datasets/deepfakesMSU/NTIRE-RobustAIGenDetection-test-public.social_media_deepfakes