AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

FunkNN: Neural Interpolation for Functional Generation

AmirEhsan Khorashadizadeh, Anadi Chaman, Valentin Debarnot, Ivan Dokmanić

ICLR 2023 · Image generation › Autoregressive and latent generation · approach: Generative latent space tooling · 2 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
Pruning LIIF-EDSR from 1.5M to 140K parametersLIIF-EDSR · CelebA-HQ 128→256 and 128→512SNR (dB): 31.6→31.6 at 256; 28.0→27.5 at 512Hurt
Patch-based FunkNN with factor training (140K params) vs. LIIF-EDSR (1.5M params)FunkNN · CelebA-HQ 128→256/512/1024SNR (dB): 32.6 vs 31.6 at 256; 28.2 vs 28.0 at 512; 24.2 vs 23.9 at 1024Mixed
Factor training strategy instead of single-scale training (ablation)FunkNN · CelebA-HQ in-distribution 128→256/512/1024SNR (dB): +1.4 at 256, +1.9 at 512, +1.7 at 1024 over singleImproved
Factor training strategy instead of single-scale training (ablation)FunkNN · LSUN-bedroom OOD 128→256SNR (dB): 26.6 vs 27.2 (-0.6)Hurt
Continuous training strategy (multiple input scales) instead of single (ablation)FunkNN · CelebA-HQ 128→512 and 128→1024SNR (dB): +1.4 at 512, +1.0 at 1024Improved
Continuous training strategy instead of single (ablation)FunkNN · CelebA-HQ 128→256 and LSUN OOD 128→256SNR (dB): -0.5 at 256; -0.8 on OODHurt
Patch-based FunkNN (factor) vs U-Net trained for 128→256Super-resolution to OOD LSUN-bedroom · LSUN-bedroom 128→256 (OOD)SNR (dB): 26.6 vs 19.9 (+6.7)Improved
U-Net super-resolution baseline vs FunkNN on inlier dataCelebA-HQ 128→256SNR (dB): 31.3 vs 32.6 for FunkNN (factor)Hurt
Patch-based FunkNN generalization to OOD data (LSUN-bedroom) vs LIIF-EDSR at 1.5M paramsFunkNN · LSUN-bedroom OOD 128→256SNR (dB): 26.6 vs 28.0Hurt
Inference speed: FunkNN vs LIIF-EDSR (1.5M)FunkNN · 100 images of 128x128 on A100time (s): 0.7 s vs 1.97 sImproved

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