AI Paper Atlas

Research map / Image generation / Editing and style transfer

MFIM: Megapixel Facial Identity Manipulation

Sanghyeon Na

ECCV 2022 · Image generation › Editing and style transfer · approach: Identity-preserving face manipulation · 4 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 style maps to baseline MFIM (ablation)MFIM baseline · CelebA-HQexpression distance: 1.116 -> 0.398Improved
add style maps to baseline MFIM (ablation)MFIM baseline · CelebA-HQpose-HN: 7.899 -> 3.795Improved
add style maps to baseline MFIM (ablation)MFIM baseline · CelebA-HQidentity distance: 70.160 -> 91.430Hurt
add shape loss L_shape to config B (ablation)MFIM · CelebA-HQshape distance: 0.823 -> 0.635Improved
add shape loss L_shape to config B (ablation)MFIM · CelebA-HQexpression distance: 0.398 -> 0.864Hurt
add pose loss L_pose to config C (ablation)MFIM · CelebA-HQpose distance: 0.085 -> 0.078Improved
add expression loss L_exp to config D (ablation)MFIM · CelebA-HQexpression distance: 0.860 -> 0.400Improved
add expression loss L_exp to config D (ablation)MFIM · CelebA-HQshape distance: 0.634 -> 0.782Hurt
add landmark loss L_lm to config B (ablation)MFIM · CelebA-HQ (30k pairs)shape distance: 0.842 -> 0.892No clear effect
use parameter loss L_param instead of landmark loss L_lm on config B (ablation)MFIM · CelebA-HQ (30k pairs)shape distance: 0.892 -> 0.842Improved
use parameter loss L_param instead of landmark loss L_lm on config B (ablation)MFIM · CelebA-HQ (30k pairs)pose distance: 0.046 -> 0.060Hurt
add partial-keypoint landmark loss L_lm to config E (ablation)MFIM · CelebA-HQpose distance: 0.057 -> 0.041Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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