AI Paper Atlas

Research map / Image generation / GANs

Wasserstein GAN With Quadratic Transport Cost

Huidong Liu, Xianfeng Gu, Dimitris Samaras

ICCV 2019 · Image generation › GANs · approach: Stabilized GAN training · 86 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 WGAN-QC instead of WGAN-divWGAN · CelebAFID: 15.2 -> 12.9Improved
use WGAN-QC instead of WGAN-divWGAN · CelebA-HQFID: 13.5 -> 7.7Improved
use WGAN-QC instead of WGAN-divWGAN · LSUN bedroomFID: 15.9 -> 13.9Improved
increase gamma from 1e-4 to 1 (ablation)WGAN-QC · CelebAFID: 23.4 -> 11.5Improved
increase gamma from 1 to 10 (ablation)WGAN-QC · CelebAFID: 11.5 -> 15.7Hurt
use WGAN-QC instead of BigGAN (supervised)unsupervised GAN · ImageNet dogInception Score: 10.39 -> 10.48Improved
use WGAN-QC instead of WGAN-div for speedWGAN · CelebAtraining time: 2.1 days -> 0.6 daysImproved
use WGAN-QC instead of WGAN-div for speedWGAN · LSUN bedroomtraining time: 2.4 days -> 1.3 daysImproved
use quadratic transport cost with OTR instead of WGAN-div regularizationWGAN · Dirac distributiongenerator convergenceImproved

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