AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection

Matic Fučka, Vitjan Zavrtanik, Danijel Skočaj

ECCV 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Industrial anomaly detection · 42 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
add normal appearance loss L_n (ablation)TransFusion · VisAimage AUROC: -31.8 when removedImproved
feed simulated previous mask estimate as input (ablation)TransFusion · VisA localizationAUPRO: -8.6 when removedImproved
add 2D positional encoding input (ablation)TransFusion · VisAimage AUROC: -1.4 when removedImproved
add mask, anomaly appearance and consistency losses (ablation)TransFusion · VisA and MVTec ADimage AUROC: about -1 each when removedImproved
combine discriminative mask with reconstruction error mask for final mask (ablation)TransFusion · VisA localizationAUPRO: -13.5 when only M_recon usedImproved
average masks over all diffusion steps instead of last mask (ablation)TransFusion · VisAimage AUROC: -1.5 when only last mask usedImproved
reduce diffusion steps from 20 to 5 (ablation)TransFusion · VisAimage AUROC: -1.0No clear effect
use linear instead of quadratic transparency schedule (ablation)TransFusion · VisAimage AUROC: -2.0 AUROC but +1.2 AUPRO for quadraticMixed
replace two-stage reconstruct-then-discriminate with single-stage transparency diffusionTransFusion vs DiffAD · VisAimage AUROC: 89.5 -> 98.5Improved
iterative diffusion inference instead of single-pass modelsTransFusion · A100 inference timeseconds per sample: 0.05 (DRAEM) -> 0.34Hurt

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