AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Wavelet Diffusion Models Are Fast and Scalable Image Generators

Hao Phung, Quan Dao, Anh Tran

CVPR 2023 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 133 citations (OpenAlex)

This paper tried a common change: Add residual connections

add frequency residual connections · CelebA-HQ 256×256 · FID: 6.25 -> 5.94 Improved

What 546 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

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
apply wavelet diffusion instead of DDGANDDGAN · CIFAR-10 32×32Time (s): 0.21 -> 0.08Improved
apply wavelet diffusion instead of DDGANDDGAN · CIFAR-10 32×32FID: 3.75 -> 4.01Hurt
use the wavelet diffusion scheme with the wavelet-embedded generatorDDGAN · STL-10 64×64FID: 21.79 -> 12.93Improved
use the wavelet diffusion scheme with the wavelet-embedded generatorDDGAN · STL-10 64×64Time (s): 0.58 -> 0.38Improved
add the wavelet-embedded generator (ablation)Wavelet Diffusion · CelebA-HQ 256×256FID: 6.55 -> 5.94Improved
add frequency residual connections (ablation)Wavelet Diffusion + W-Generator · CelebA-HQ 256×256FID: 6.25 -> 5.94Improved
add frequency-aware upsampling and downsampling blocks (ablation)Wavelet Diffusion + W-Generator · CelebA-HQ 256×256FID: 6.23 -> 5.94Improved
add frequency-aware upsampling and downsampling blocks (ablation)Wavelet Diffusion + W-Generator · CelebA-HQ 256×256Time (s): 0.61 -> 0.79Hurt
add frequency bottleneck blocks (ablation)Wavelet Diffusion + W-Generator · CelebA-HQ 256×256FID: 6.18 -> 5.94Improved
add the reconstruction term (ablation)Wavelet Diffusion · CelebA-HQ 256×256FID: around 0.6 points to 5.94Improved
process both low- and high-frequency bottleneck subbands instead of passing high-frequency subbands through (ablation)Wavelet Diffusion + W-Generator · STL-10 64×64FID: 12.96 vs. 12.93No clear effect
use progressive two-level upsampling (ablation)Wavelet Diffusion · CelebA-HQ 256×256FID: 13.11 vs. 5.94Hurt

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