AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

ALRe: Outlier Detection for Guided Refinement

Mingzhu Zhu, Zhang Gao, Junzhi Yu, Bingwei He, Jiantao Liu

ECCV 2020 · Image generation › Autoregressive and latent generation · approach: Reconstruction-based diffusion image detection · 2 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 ALRe instead of weighted-median-filter change as outlier detectoroutlier detection · Synthetic Middlebury 2014, M=50IoU: 0.814 -> 0.978Improved
use ALRe instead of weighted-median-filter change as outlier detectoroutlier detection · Synthetic Middlebury 2014, M=100IoU: 0.727 -> 0.958Improved
use ALRe instead of weighted-median-filter change as outlier detectoroutlier detection · Synthetic Middlebury 2014, M=200IoU: 0.556 -> 0.868Improved
increase number of outlier regions M from 50 to 200 (ablation)ALRe · Synthetic Middlebury 2014IoU: 0.978 -> 0.868Hurt
increase number of outlier regions M from 50 to 200 (ablation)WMF detection · Synthetic Middlebury 2014IoU: 0.814 -> 0.556Hurt
increase ALRe expectation (larger box-filter R) (ablation)ALRe and WMF · Synthetic Middlebury 2014IoUNo clear effect
use ALRe weights in WMF for depth refinementWMF · RGB-D depth map refinement (qualitative)visual qualityImproved
use ALRe weights in GF for featheringGF · guided feathering (qualitative)visual qualityImproved
use ALRe weights instead of cross check for disparity refinementWLS/JBF · Middlebury 2003 disparity (qualitative)visual qualityNo clear effect
use ALRe weights instead of model-based weights for transmission refinementWLS · haze removal (qualitative)visual qualityNo clear effect

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