AI Paper Atlas

Research map / Image generation / GANs

Boundless: Generative Adversarial Networks for Image Extension

Piotr Teterwak, Aaron Sarna, Dilip Krishnan, Aaron Maschinot, David Belanger, Ce Liu, William T. Freeman

ICCV 2019 · Image generation › GANs · approach: Large-scale GAN latent analysis · 115 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 semantic (InceptionV3 activation) conditioning to the discriminator (ablation)Boundless · Places365 top-50 classes, 25% extensionFID: 0.85 -> 0.79Improved
add semantic conditioning to the discriminator (ablation)Boundless · Places365 top-50 classes, 50% extensionFID: 5.01 -> 3.46Improved
add semantic conditioning to the discriminator (ablation)Boundless · Places365 top-50 classes, 75% extensionFID: 19.17 -> 8.79Improved
add semantic conditioning to the discriminator (ablation)Boundless · Places365 top-50 classes, central inpaintingFID: 2.73 -> 2.53Improved
add semantic conditioning to the discriminator (ablation)Boundless · Places365 top-50 classes, 75% extensionPSNR: 9.08 -> 8.07Hurt
replace Boundless conditioning with a perceptual lossBoundless · Places365 top-50 classes, 75% extensionFID: 8.79 -> 14.15Hurt
replace Boundless conditioning with feature matchingBoundless · Places365 top-50 classes, 75% extensionFID: 8.79 -> 14.74Hurt
replace Boundless conditioning with a perceptual lossBoundless · Places365 top-50 classes, 25% extensionFID: 0.79 -> 0.40Improved
replace Boundless conditioning with a perceptual lossBoundless · Places365 top-50 classes, 50% extensionFID: 3.46 -> 2.32Improved
use Boundless instead of DeepFill with contextual attention and refinement networkBoundless · Places365 top-50 classes, 75% extensionFID: 31.21 -> 8.79Improved
remove skip connections in the generator (ablation)Boundless · Places365 qualitative ablationvisual qualityImproved
add instance normalization to the generator (ablation)Boundless · Places365 qualitative ablationvisual qualityImproved

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