AI Paper Atlas

Research map / 3D vision / Face and shape

Fake It Till You Make It: Face Analysis in the Wild Using Synthetic Data Alone

Erroll Wood, Tadas Baltrušaitis, Charlie Hewitt, Sebastian Dziadzio, Thomas J. Cashman, Jamie Shotton

ICCV 2021 · 3D vision › Face and shape · approach: 3D morphable face model regression · 286 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 full augmentation (appearance and geometry) to synthetic training (ablation)Ours (synthetic) · 300W CommonNME: 4.25 -> 3.09Improved
add geometric augmentation on top of appearance augmentation (ablation)Ours (synthetic) · 300W ChallengingNME: 6.80 -> 4.86Improved
add label adaptation to synthetic-trained model (ablation)Ours (synthetic) · 300W CommonNME: 5.61 -> 3.09Improved
add label adaptation to synthetic-trained model (ablation)Ours (synthetic) · 300W PrivateFR10%: 4.67 -> 0.50Improved
add label adaptation to real-data-trained model (ablation)Ours (real) · 300W CommonNME: 3.44 -> 3.37No clear effect
add hair and clothing to synthetic faces (ablation)Ours (synthetic) · 300W CommonNME: 3.36 -> 3.09Improved
add hair and clothing to synthetic faces (ablation)Ours (synthetic) · 300W PrivateFR10%: 2.17 -> 0.50Improved
add clothing to synthetic faces (ablation)Ours (synthetic) · 300W ChallengingNME: 5.09 -> 4.86Mixed
add hair on top of clothing-free faces (ablation)Ours (synthetic) · 300W ChallengingNME: 5.37 -> 5.09Mixed
train on synthetic instead of real datalandmark localization ResNet34 · 300W ChallengingNME: 5.77 -> 4.86Improved
train on synthetic instead of real dataface parsing UNet · Helen* face parsing (overall)F1: 91.6 -> 92.0Improved
train on synthetic instead of real dataface parsing UNet · LaPa face parsing (mean)F1: 90.9 -> 90.1Hurt

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