AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Procedural Image Programs for Representation Learning

Manel Baradad, Chun-Fu Chen, Jonas Wulff, Tongzhou Wang, Rogerio Feris, Antonio Torralba, Phillip Isola

NeurIPS 2022 · Image generation › Autoregressive and latent generation · approach: Classical generative image synthesis · 5 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
mix images from six different shaders with MixUp (ablation)MoCo v2 pre-training on Shaders-21k · ImageNet-1ktop-1 accuracy: 30.25 -> 44.83Improved
mix images from six different shaders with MixUp (ablation)MoCo v2 pre-training on Shaders-21k · VTAB Structuredtop-1 accuracy: 32.85 -> 31.84Hurt
use MixUp rather than original shader samples (ablation)MoCo v2 pre-training on StyleGAN Oriented · ImageNet-1ktop-1 accuracy: 38.12 -> 31.73Hurt
render new shader samples during each training batch (ablation)S-21k MixUp · ImageNet-1ktop-1 accuracy: 44.83 -> 45.25Improved
use supervised contrastive learning rather than SimCLR (ablation)Shaders-21k pre-training · ImageNet-100top-1 accuracy: 47.74 -> 48.10No clear effect
increase the number of shader programs (ablation)representation learning on Shaders-21k · ImageNet-100top-1 accuracyImproved
select curated TwiGL shaders rather than a random same-size subset of Shaders-21k (ablation)SimCLR pre-training · ImageNet-100top-1 accuracy: 42.26 -> 43.26Improved
increase MixUp from one raw sample to five mixed samples (ablation)S-21k image generation · ImageNet-100 FIDFID: 41.00 -> 35.03Improved
generate interpolated images with StyleGAN rather than MixUp (ablation)S-21k image generation · ImageNet-1ktop-1 accuracy: 44.83 -> 35.19Hurt
add MixUp to StyleGAN-generated S-21k images (ablation)S-21k StyleGAN · ImageNet-1ktop-1 accuracy: 35.19 -> 36.46Improved
use supervised classification instead of unsupervised SimCLR (ablation)representation learning with shader programs · ImageNet-100top-1 accuracy: supervision helps with more than 20 and fewer than 1000 shaders; not outside this regimeMixed

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