AI Paper Atlas

Research map / Image generation / Editing and style transfer

Detecting Photoshopped Faces by Scripting Photoshop

Sheng-Yu Wang, Oliver Wang, Andrew Owens, Richard Zhang, Alexei A. Efros

ICCV 2019 · Image generation › Editing and style transfer · approach: Localized and attribute-specific editing · 147 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
add augmentation (resizing, JPEG, color) to low-res classifier (ablation)Low-res global classifier · Validation (random FAL)accuracy (%): 97.0 -> 93.7Hurt
add augmentation to low-res classifier after Facebook post-processing (ablation)Low-res global classifier · Artist test set, Facebook post-processingaccuracy (%): 57.0 -> 67.0Improved
increase input resolution from 400 to 700Global classifier with aug. · Validation (random FAL)accuracy (%): 93.7 -> 97.1Improved
increase input resolution to 700 on Facebook post-processed imagesGlobal classifier with aug. · Artist test set, Facebook post-processingaccuracy (%): 67.0 -> 55.0Hurt
add reconstruction loss to EPE and multiscale gradient losses (ablation)Local flow prediction model · Val (Rand-FAL)ΔPSNR (dB): +2.38 -> +2.69Improved
add reconstruction loss on top of MultiG for Artist-Liquify (ablation)Local flow prediction model · Artist-LiquifyΔPSNR (dB): -0.84 -> -0.72Improved
add multiscale gradient and reconstruction losses to EPE-only (ablation)Local flow prediction model · Val (Rand-FAL)EPE (lower better): 0.51 -> 0.52Mixed
train on random FAL, test on artist-made FALHigh-res global classifier · Validation vs Artist testaccuracy (%): 97.1 -> 90.0Hurt
test on generic Liquify instead of Face-Aware LiquifyGlobal classifier · Artist test setaccuracy (%): 90.0 -> 64.0Hurt
test on out-of-distribution warps (X2face flows) instead of FAL on facesGlobal classifier · Generalizationaccuracy (%): 93.7 -> 64.7Hurt
print and rescan images (rebroadcasting)Global classifier · Robustnessaccuracy (%): 94.2 -> 69.2Hurt

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