AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

∞-Brush: Controllable Large Image Synthesis with Diffusion Models in Infinite Dimensions

Minh-Quan Le, Alexandros Graikos, Srikar Yellapragada, Rajarsi Gupta, Joel Saltz, Dimitris Samaras

ECCV 2024 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis

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 cross-attention neural operator instead of vanilla grid cross-attention (ablation)∞-Brush · TCGA-BRCA 1.25x 4096x4096CLIP FID: 3.81 -> 2.63Improved
add cross-attention neural operator instead of vanilla grid cross-attention (ablation)∞-Brush · TCGA-BRCA 1.25x 4096x4096Crop FID: 16.28 -> 14.76Improved
use function-space conditional diffusion instead of patch-based generationvery large image synthesis · TCGA-BRCA 1.25x 4096x4096CLIP FID: 2.75 -> 2.63Improved
use function-space conditional diffusion instead of patch-based generationvery large image synthesis · TCGA-BRCA 1.25x 4096x4096Crop FID: 11.30 -> 14.76Hurt
use ∞-Brush instead of SDXLlarge image synthesis · TCGA-BRCA 5x 1024x1024CLIP FID: 6.64 -> 3.74Improved
use ∞-Brush instead of SDXLlarge image synthesis · TCGA-BRCA 5x 1024x1024Crop FID: 6.98 -> 17.87Hurt
use ∞-Brush instead of patch-based methodlarge image synthesis · NAIP 1024x1024Crop FID: 43.76 -> 48.65Hurt
add conditioning to infinite-dimensional diffusion model∞-Diff · CelebA-HQCLIP FID: 9.44 -> 8.38Improved
train on 0.4% pixel subsets of full images instead of full-resolution or patch training∞-Brush · training costepoch time (hr): 12 vs 300 (patch-based at 4096)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