AI Paper Atlas

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

Unsupervised Foreground Extraction via Deep Region Competition

Peiyu Yu, Sirui Xie, Xiaojian Ma, Yixin Zhu, Ying Nian Wu, Song-Chun Zhu

NeurIPS 2021 · Detection and segmentation › Open-set and few-shot detection · approach: Unsupervised object and part discovery · 4 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 pseudo label (ablation)DRC · BirdsIoU: 48.7 -> 56.4Improved
add TV-norm regularization (ablation)DRC · BirdsIoU: 53.0 -> 56.4Improved
add orthogonality regularization (ablation)DRC · BirdsIoU: 54.3 -> 56.4Improved
use persistent chain instead of short-run MCMC (ablation)DRC · BirdsIoU: 52.5 -> 56.4Improved
use latent-space EBM instead of amortized inference networks (ablation)DRC · BirdsIoUImproved
DRC vs ReDOReDO · BirdsIoU: 46.5 -> 56.4Improved
DRC vs Slot-AttentionSlot-Attention · TM-dSpritesIoU: 7.3 -> 78.8Improved
DRC vs Slot-AttentionSlot-Attention · CLEVR6IoU: 83.6 -> 84.7No clear effect
DRC vs OneGAN weakly supervisedOneGAN · BirdsIoU: 55.5 -> 56.4No clear effect
DRC vs GrabCut on CarsGrabCut · CarsIoU: 61.3 -> 72.4Improved
evaluate on held-out test split instead of trainDRC · BirdsIoU: 54.8 -> 54.6No clear effect
train on Dogs, test on unseen BirdsDRC vs ReDO · BirdsIoU: 22.2 -> 41.3Improved

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