AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Discovering Transferable Forensic Features for CNN-Generated Images Detection

Keshigeyan Chandrasegaran, Ngoc-Trung Tran, Alexander Binder, Ngai-Man Cheung

ECCV 2022 · Image generation › Autoregressive and latent generation · approach: Generalizable fake image detection · 36 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
dropout of top-k forensic features chosen by FF-RS rather than random-k (ablation)ResNet-50 universal detector · ForenSynths StyleGAN2AP: 55.3 vs 98.6 (random-k)Improved
drop top-k T-FF feature maps (ablation)ResNet-50 universal detector · ForenSynths ProGANGAN detection accuracy: 100. -> 3.2Hurt
drop top-k T-FF feature maps (ablation)EfficientNet-B0 universal detector · ForenSynths ProGANAP: 100. -> 50.0Hurt
drop low-k feature maps (ablation)ResNet-50 universal detector · ForenSynths StyleGAN2AP: 99.1 -> 99.1No clear effect
increase dropout size from k=29 to k=228 (ablation)ResNet-50 universal detector · ForenSynths ProGANAP: 98.6 -> 58.6Hurt
convert counterfeits to grayscale at test time (ablation)ResNet-50 universal detector · ForenSynths BigGANGAN detection accuracy: 85.1 -> 48.8Hurt
convert counterfeits to grayscale at test time (ablation)ResNet-50 universal detector · ForenSynths ProGANGAN detection accuracy: 100.0 -> 81.5Hurt
convert counterfeits to grayscale at test time (ablation)universal detector · ForenSynths unseen GANsmedian counterfeit probability: drop by more than 89%Hurt
add random grayscale (50% probability) augmentation during training (CR-universal)ResNet-50 universal detector · ProGAN, Mood median test% color-conditional T-FF: 85.1 -> 55.3Improved
add random grayscale augmentation during training (CR-universal)EfficientNet-B0 universal detector · ProGAN, Mood median test% color-conditional T-FF: 51.9 -> 20.0Improved
add random grayscale augmentation during training (CR-universal)universal detector · ForenSynths unseen GANsrobustness to color ablationImproved
use pixel-wise attribution (Guided-GradCAM, LRP) to find forensic featuresuniversal detector · ProGAN-trained detectorqualitative discovery of T-FFNo clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy