AI Paper Atlas

Research map / Image generation / GANs

Generative Semantic Manipulation with Mask-Contrasting GAN

Xiaodan Liang, Hao Zhang, Liang Lin, Eric Xing

ECCV 2018 · Image generation › GANs · approach: Large-scale GAN latent analysis · 70 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 cycle-consistency loss to contrast-only objective (ablation)Contrast-GAN · Cityscapes labels->photosper-pixel accuracy: 0.53 -> 0.57Improved
add LSGAN classification loss to contrast + cycle (ablation)Contrast-GAN · Cityscapes labels->photosper-pixel accuracy: 0.57 -> 0.58Improved
add LSGAN classification loss to contrast + cycle (ablation)Contrast-GAN · Cityscapes labels->photosclass IoU: 0.13 -> 0.16Improved
add LSGAN classification loss to contrast + cycle (ablation)Contrast-GAN · Cityscapes labels->photosper-class accuracy: 0.22 -> 0.21No clear effect
add cycle-consistency loss to contrast + classify (ablation)Contrast-GAN · Cityscapes photos->labelsper-pixel accuracy: 0.55 -> 0.61Improved
add LSGAN classification loss to contrast + cycle (ablation)Contrast-GAN · Cityscapes photos->labelsclass IoU: 0.15 -> 0.18Improved
add the contrasting objective to cycle-consistency (CycleGAN to full model) (ablation)CycleGAN · Cityscapes labels->photosper-pixel accuracy: 0.52 -> 0.58Improved
add the contrasting objective to cycle-consistency (CycleGAN to full model) (ablation)CycleGAN · Cityscapes photos->labelsper-pixel accuracy: 0.58 -> 0.61Improved
use contrast objective alone instead of CycleGANContrast-GAN · Cityscapes photos->labelsper-pixel accuracy: 0.58 -> 0.54Hurt
use one shared conditional generator instead of separate generators per semantic (ablation)Contrast-GAN · Cityscapes labels->photosper-pixel accuracy: 0.57 -> 0.58No clear effect
use one shared conditional generator instead of separate generators per semantic (ablation)Contrast-GAN · Cityscapes photos->labelsclass IoU: 0.17 -> 0.18No clear effect
add auxiliary global discriminator D_I (ablation)Mask Contrast-GAN · MSCOCO cat->dog (AMT)% labeled real: 4.3% -> 4.8%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