atlas

Research map / Image generation / Editing and style transfer

Cross-Modal Deep Face Normals With Deactivable Skip Connections

CVPR 2020 · Image generation › Editing and style transfer · approach: Identity-preserving face manipulation · 26 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Use cross-modal learning with image and normal encoders/decoders and deactivable skip connectionsFace normal estimation · Photofacemean angular error (degrees): 24.8$\pm$6.8 -> 22.8$\pm$6.5Helped
Use cross-modal learning with image and normal encoders/decoders and deactivable skip connectionsFace normal estimation · Florencemean angular error (degrees): 14.1$\pm$2.16 -> 11.3$\pm$1.5Helped
Fine-tune on PhotofaceOurs · Photofacemean angular error (degrees): 22.8$\pm$6.5 -> 12.0$\pm$5.3Helped
Remove skip connections (ablation)Cross-modal face normal model · Photoface and Florencenormal estimation accuracyHurt
Remove the normal encoder (ablation)Cross-modal face normal model · Photoface and Florencenormal estimation accuracyHurt
Remove the normal encoder and image decoder (ablation)Cross-modal face normal model · In-the-wild face imagesnormal estimation accuracyHurt

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