AI Paper Atlas

Research map / 3D vision / Face and shape

3D-FM GAN: Towards 3D-Controllable Face Manipulation

Yuchen Liu, Zhixin Shu, Yijun Li, Zhe Lin, Richard Zhang, S.Y. Kung

ECCV 2022 · 3D vision › Face and shape · approach: Face reflectance and relightable rendering · 18 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
encode photo into W+ instead of render into W (ablation)exclusive modulation · FFHQ held-out 5k, manipulationidentity preservation (Id): 0.57 -> 0.66Improved
encode photo into W+ instead of render into W (ablation)exclusive modulation · FFHQ held-out 5k, manipulationlandmark error (LM): 17.8 -> 27.3Hurt
encode render into W+ instead of W (ablation)exclusive modulation · FFHQ held-out 5k, manipulationFID: 14.5 -> 25.8Hurt
use multiplicative instead of concatenation co-modulation (ablation)2-encoder co-modulation · FFHQ held-out 5k, manipulationFID: 17.8 -> 12.4Improved
use multiplicative instead of concatenation co-modulation (ablation)2-encoder co-modulation · FFHQ held-out 5k, manipulationlandmark error (LM): 59.3 -> 22.2Improved
use multiplicative instead of tensor transform co-modulation (ablation)2-encoder co-modulation · FFHQ held-out 5k, manipulationFID: 18.6 -> 12.4Improved
add third encoder to multiplicative co-modulation (ablation)co-modulation · FFHQ held-out 5k, manipulationlandmark error (LM): 22.2 -> 17.2Improved
add third encoder to multiplicative co-modulation (ablation)co-modulation · FFHQ held-out 5k, manipulationface content similarity error (FC): 0.025 -> 0.020Improved
add third encoder to multiplicative co-modulation (ablation)co-modulation · FFHQ held-out 5k, manipulationidentity preservation (Id): 0.66 -> 0.66No clear effect
alternate reconstruction and disentanglement training instead of either alone (ablation)Render-W · synthetic data, 140k iterationsidentity and editabilityImproved
add real-data (FFHQ) reconstruction to training (ablation)Render-W · 280k iterationsphoto-realismImproved
two-phase (synthetic then real) instead of real-only reconstruction (ablation)Render-W · 280k iterationsidentity preservationImproved

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