AI Paper Atlas

Research map / Image generation / GANs

Neural Characteristic Function Learning for Conditional Image Generation

Shengxi Li, Jialu Zhang, Yifei Li, Mai Xu, Xin Deng, Li Li

ICCV 2023 · Image generation › GANs · approach: Stabilized GAN training · 6 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 an auxiliary classifier to estimate p(y|x) instead of ground-truth labels (ablation)CCF-GAN · CIFAR10FID: 7.48 -> 6.08Improved
use fixed linear ty in [-1, 1] rather than uniform or Gaussian random ty (ablation)CCF-GAN · CIFAR10FID: 6.08 vs 7.05 (uniform), 7.07 (Gaussian)Improved
vary fixed ty range alpha (0.5, 1, 10) (ablation)CCF-GAN · CIFAR10FID: 6.51 / 6.08 / 7.02Mixed
increase number of projections k via NCF network (ablation)CCF-GAN · CIFAR10FIDImproved
replace cross-entropy cGAN discriminator with neural characteristic function discriminatorBigGAN architecture · ImageNet 128x128FID / IS: FID 22.77 -> 11.34 (vs BigGAN); IS 180.84 vs 68.27 (ReACGAN)Improved
replace cross-entropy cGAN discriminator with neural characteristic function discriminatorBigGAN architecture · VGGFace_c1000FID: 24.07 -> 5.70Improved
use CF-based discrepancy to reduce mode collapse and training instabilityCCF-GAN · CIFAR10 / VGGFace_c1000 (precision-recall); 48 training conditionsprecision, recall, successful trialsImproved

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