AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Foreground-Background Separation through Concept Distillation from Generative Image Foundation Models

Mischa Dombrowski, Hadrien Reynaud, Matthew Baugh, Bernhard Kainz

ICCV 2023 · Image generation › Fast diffusion sampling · approach: Diffusion models as perception tools · 9 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
train a U-Net on refined masks of original data plus synthetic data D'_s (ablation)Concept distillation pipeline · CUBIoU: 75.1 -> 77.2Improved
train a U-Net on preliminary masks instead of using them directly (ablation)Preliminary masks · CUBIoU: 29.0 -> 66.5Improved
replace simple crop inpainting with fine-tuned diffusion inpainting for mask refinement (ablation)Mask refinement · CUBIoU: 30.7 -> 63.6Improved
use refined masks directly instead of U-Net trained on preliminary masks (ablation)Pipeline · CUBIoU: 66.5 -> 63.6No clear effect
average attention over more reverse diffusion steps up to T0 about 40 (ablation)Mean attention maps · CUBAUC-ROCImproved
include higher (noisier) diffusion steps beyond T0 about 40 (ablation)Mean attention maps · CUBAUC-ROCHurt
instance-wise normalization of attention maps (ablation)Mean attention maps · CUBAUC-ROC: 97.08 (no normalization) vs 97.1No clear effect
Monte Carlo averaging over repeated diffusion samples N (ablation)Preliminary masks · CUBAUC-ROCNo clear effect
increase classifier-free guidance scale toward foreground (ablation)Fine-tuned LDM · CUBFID: 9.8 -> 22.3Hurt
use object token only or list of objects as prompt rather than 'a photo of a bird' (ablation)Preliminary masks · CUBAUC-ROCHurt
replace Layered GAN with diffusion concept distillationForeground-background synthesis · CUB / Stanford CarsFID: 12.9 -> 9.8; 19.0 -> 13.4Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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