AI Paper Atlas

Research map / Image generation / GANs

Positional Encoding As Spatial Inductive Bias in GANs

Rui Xu, Xintao Wang, Kai Chen, Bolei Zhou, Chen Change Loy

CVPR 2021 · Image generation › GANs · approach: Large-scale GAN latent analysis · 71 citations (OpenAlex)

Check 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
Remove zero padding (ablation)StyleGAN2 · FFHQFID@256, 20M: 5.79 -> 6.45Hurt
Remove zero padding (ablation)PGGAN · CelebA cropped 128x128SWD Avg. (× 10^3): 4.149 -> 8.534Hurt
Add SPE to the padding-free PGGAN (ablation)PGGAN · CelebA cropped 128x128SWD Avg. (× 10^3): 8.534 -> 5.694Improved
Remove zero padding (ablation)DCGAN · LSUN Bedroom 128x128SWD Avg. (×10^3): 19.07 -> 31.24Hurt
Add SPE to the padding-free DCGAN (ablation)DCGAN · LSUN Bedroom 128x128SWD Avg. (×10^3): 31.24 -> 17.93Improved
Add SPE to the padding-free PGGAN (ablation)PGGAN · LSUN Bedroom 128x128SWD Avg. (× 10^3): 10.081 -> 5.371Improved
Use expanded SPE in multi-scale trainingStyleGAN2-C2-256 · FFHQFID@512, 25M: 3.41 -> 2.93Improved
Use expanded rather than interpolated SPE (ablation)MS-PIE StyleGAN2-C2-256 · FFHQFID@512, 25M: 3.15 -> 2.93Improved
Use Cartesian grid rather than expanded SPE (ablation)MS-PIE StyleGAN2-C2-256 · FFHQFID@512, 25M: 2.93 -> 3.50Hurt
Reduce channel multiplier from two to one (ablation)MS-PIE SPE-expand · FFHQFID@512, 25M: 2.93 -> 3.40Hurt
Use a learnable constant input instead of expanded SPE (ablation)MS-PIE StyleGAN2-C2-256 · FFHQFID@512, 25M: 2.93 -> 3.35Hurt
Use padding-free training rather than retaining zero padding (ablation)MS-PIE StyleGAN2-C2-256 · FFHQFID@512, 25M: 2.93 -> 4.17Hurt

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