AI Paper Atlas

Research map / Image generation / GANs

InterFaceGAN: Interpreting the Disentangled Face Representation Learned by GANs

Yujun Shen, Ceyuan Yang, Xiaoou Tang, Bolei Zhou

TPAMI 2020 · Image generation › GANs · approach: GAN-based image manipulation · 576 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
use StyleGAN W space instead of Z space for boundary correlation (ablation)InterFaceGAN on StyleGAN · FF-HQ boundary correlation (Age-Glasses)cosine similarity: 0.72 -> 0.05Improved
add conditional manipulation (subspace projection) (ablation)InterFaceGAN on PGGAN · PGGAN attribute editingqualitative entanglementImproved
use W space instead of Z space for re-scoring Age manipulation (ablation)InterFaceGAN on StyleGAN · Re-scoring, Glasses score when editing Agescore change: 0.32 -> 0.20Improved
edit only layers 08-17 instead of early layers (ablation)InterFaceGAN on StyleGAN W · Layer-wise Pose re-scoringscore change: 0.42 -> 0.00Hurt
edit only layers 02-03 for gender (ablation)InterFaceGAN on StyleGAN W · Layer-wise Gender re-scoringscore change: 0.45 -> 0.44Mixed
use StyleGAN W space instead of PGGAN Z space for identity preservation (Gender) (ablation)InterFaceGAN · Identity discrepancy (Gender)identity discrepancy: 0.61 -> 0.40Improved
edit smile instead of genderInterFaceGAN on PGGAN · Identity discrepancyidentity discrepancy: 0.61 -> 0.24Improved
use W space instead of Z space for linear separability (ablation)InterFaceGAN on StyleGAN · boundary classification accuracyaccuracyImproved
use optimization-based instead of encoder-based GAN inversionInterFaceGAN real editing · inversion time and identitytime per image: 3 minutes vs 0.1 secondsHurt
move latent code far from the boundaryInterFaceGAN on PGGAN · gender editingidentity preservationHurt
train pix2pixHD on synthetic pairs for smilepix2pixHD with InterFaceGAN data · Real face editingqualitativeHurt

Later papers that compared with this method

16 later accepted papers report InterfaceGAN as a baseline in their results tables; 14 of them report a better number on most of the results they share with it.

8 more not listed.

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