AI Paper Atlas

Research map / Image generation / GANs

Learning Continuous Face Age Progression: A Pyramid of GANs

Hongyu Yang, Di Huang, Yunhong Wang, Anil K. Jain

TPAMI 2019 · Image generation › GANs · approach: GAN-based image manipulation · 55 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 pyramid-structured discriminator over one-pathway discriminator (ablation)age progression GAN · CACDmean absolute age error (years): 2.48 -> 0.99Improved
add pyramid-structured discriminator over one-pathway discriminator (ablation)age progression GAN · MORPHmean absolute age error (years): 2.67 -> 0.85Improved
add pyramid discriminator (ablation)age progression GAN · CACDverification confidence, aged face 3: 87.43 -> 91.22Improved
add pyramid discriminator (ablation)age progression GAN · MORPHverification confidence, aged face 3: 86.94 -> 88.12Improved
add pyramid discriminator (ablation)age progression GAN · CACDverification confidence, aged face 1: 95.51 -> 94.61Hurt
use k class-wise discriminators instead of single discriminator with k logits (ablation)age progression GAN · qualitative facesvisual aging qualityImproved
add auxiliary age classifier (AC-GAN) to discriminator (ablation)age progression GAN · qualitative facesaging effect naturalnessNo clear effect
independently train per age cluster models (previous work) vs joint conditional modelage progression GAN · CACDverification rate, aged face 3 (%): 99.11% -> 99.98Improved
independently train per age cluster models (previous work) vs joint conditional modelage progression GAN · MORPHverification rate, aged face 3 (%): 95.94% -> 99.88Improved
compare proposed method with prior work in pairwise human studyage progression GAN · 138 image pairs from published paperspreference votes (%): 71.74% vs 19.28%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