AI Paper Atlas

Research map / Image generation / GANs

Image Manipulation with Perceptual Discriminators

Diana Sungatullina, Egor Zakharov, Dmitry Ulyanov, Victor Lempitsky

ECCV 2018 · Image generation › GANs · approach: GAN-based image manipulation · 27 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 perceptual VGG features to the discriminator (VGG-GAN vs DCGAN)DCGAN · CelebA C2ST smileC2ST cross-entropy x10^-2: 0.6 -> 8.6Improved
add perceptual VGG features to the discriminator (VGG-GAN vs DCGAN)DCGAN · CelebA C2ST hair colorC2ST cross-entropy x10^-2: 0.06 -> 0.96Improved
add perceptual VGG features to the discriminator (VGG-GAN vs DCGAN)DCGAN · CelebA classification loss genderlog-loss: 1.5 -> 0.1Improved
add perceptual VGG features to the discriminator (VGG-GAN vs DCGAN)DCGAN · CelebA classification loss hair colorlog-loss: 2.33 -> 1.3Improved
use VGG* (leaky ReLU, average pooling) instead of VGG as feature network (ablation)VGG-GAN · CelebA C2ST hair colorC2ST cross-entropy x10^-2: 0.96 -> 1.29Improved
use VGG* (leaky ReLU, average pooling) instead of VGG as feature network (ablation)VGG-GAN · CelebA C2ST smileC2ST cross-entropy x10^-2: 8.6 -> 5.2Hurt
use VGG* instead of VGG as feature network (ablation)VGG-GAN · CelebA qualitativeimage qualityNo clear effect
pretrained weights instead of random weights in the perceptual feature network (ablation)MS-CycleGAN on CelebA-HQ 256 · CelebA-HQ male-femalequalitative realismImproved
perceptual discriminator (VGG*-MS-CycleGAN) vs CycleGANCycleGAN · apple to orangequalitative realismImproved
perceptual discriminator vs CycleGANCycleGAN · Monet-photo style transferqualitative realismNo clear effect
perceptual discriminator vs additive perceptual lossstandard discriminator · aligned image translationimage qualityNo clear effect
naive perceptual discriminatorunconditional generation · unconditional image generationtraining stabilityHurt

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