AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Accelerating Image Generation with Sub-path Linear Approximation Model

Chen Xu, Tianhui Song, Weixin Feng, Xubin Li, Tiezheng Ge, Bo Zheng, Limin Wang

ECCV 2024 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 14 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
use sub-path linear approximation (SPLAM) instead of consistency distillation (LCM)SDv1.5 distillation, 4 steps · MSCOCO2014-30k (w=8)FID: 14.53 -> 13.39Improved
use SPLAM instead of LCM at 2 stepsSDv1.5 distillation · MSCOCO2014-30k (w=3)FID: 14.29 -> 12.31Improved
use SPLAM instead of LCM at 4 stepsSDv1.5 distillation · MSCOCO2017-5k (w=3)FID: 21.41 -> 20.77Improved
use SPLAM instead of LCM with SDv2.1-base on LAION test setSDv2.1-base distillation · LAION-Aesthetics-6+ (1 step)FID: 32.41 -> 32.64No clear effect
use SPLAM instead of LCM with SDv2.1-base on COCO-30k zero-shotSDv2.1-base distillation · COCO-30k (1 step)FID: 43.87 -> 40.52Improved
add multiple estimation strategy to allow large skipping step size k (ablation)SPLAM · SDv1.5, w=8, skipping step size ablationFIDImproved
increase skipping step size k without sub-path linear ODE (LCM) (ablation)LCM · SDv1.5, w=8, skipping step size ablationFIDHurt
raise skipping step size from 20 to 100 with multiple estimation (ablation)SPLAM · SDv1.5 training costtraining cost: about 6 A100 GPU days vs about 16 for LCMImproved
use sub-path linear ODE objective instead of LCM objective (ablation)SPLAM · COCO samples, consistency mapping error deltamapping error: about 10% reduction at t=800Improved
increase guidance scale w (ablation)SPLAM · SDv1.5FID / CLIP ScoreMixed

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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