AI Paper Atlas

Research map / Image generation / Editing and style transfer

MarioNETte: Few-Shot Face Reenactment Preserving Identity of Unseen Targets

Sungjoo Ha, Martin Kersner, Beomsu Kim, Seokjun Seo, Dongyoung Kim

AAAI 2020 · Image generation › Editing and style transfer · approach: Identity-preserving face manipulation · 161 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 the landmark transformer (ablation)MarioNETte · CelebV different-identity reenactment, 1 targetCSIM: 0.520 -> 0.568Improved
Add the landmark transformer (ablation)MarioNETte · CelebV different-identity reenactment, 8 targetsCSIM: 0.608 -> 0.661Improved
Add the landmark transformer (ablation)MarioNETte · CelebV different-identity reenactmentPRMSE: 3.41 -> 3.70; 3.26 -> 3.57Hurt
Replace AdaIN with image attention (ablation)MarioNETte · CelebV different-identity reenactment, 1 targetCSIM: 0.063 -> 0.333Improved
Replace AdaIN with image attention (ablation)MarioNETte · CelebV different-identity reenactment, 8 targetsCSIM: 0.069 -> 0.472Improved
Add target feature alignment instead of using attention alone (ablation)MarioNETte · CelebV different-identity reenactment, 1 targetCSIM: 0.333 -> 0.530Improved
Add target feature alignment instead of using attention alone (ablation)MarioNETte · CelebV different-identity reenactment, 8 targetsCSIM: 0.472 -> 0.605Improved
Use alignment without image attention (ablation)MarioNETte · CelebV different-identity reenactmentPRMSE: 3.44; 3.27Hurt
Combine image attention and target feature alignment (ablation)MarioNETte · CelebV different-identity reenactmentCSIM: 0.530 -> 0.520; 0.605 -> 0.608Mixed
Add image attention instead of AdaIN (ablation)MarioNETte · VoxCeleb1 self-reenactment, 1 targetCSIM: 0.183 -> 0.611Improved
Use target feature alignment only instead of image attention only (ablation)MarioNETte · VoxCeleb1 self-reenactment, 8 targetsPRMSE: 2.909 -> 2.563Improved
Add image attention and target feature alignment (ablation)MarioNETte · CelebV different-identity reenactmentidentity preservation (CSIM)Improved

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