AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

Novel Class Discovery in Semantic Segmentation

Yuyang Zhao, Zhun Zhong, Nicu Sebe, Gim Hee Lee

CVPR 2022 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Consistency regularization for segmentation · 32 citations (OpenAlex)

Check 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 over-clustering instead of exact-clustering (ablation)Basic framework · PASCAL-5^inovel-class mIoU: 4.7% improvement in average mIoU; Fold0 1.6%; Fold1 -1.88%; Fold2 9.00%; Fold3 10.09%Mixed
add entropy ranking and train with the clean split (ablation)Basic with over-clustering · PASCAL-5^inovel-class mIoU (AVG): 54.51 vs 54.36No clear effect
add dynamic reassignment of clean data (ablation)Over-clustering + entropy ranking · PASCAL-5^inovel-class mIoU (AVG): 0.78% improvement in average mIoUImproved
add self-training on unclean data (ablation)Over-clustering + entropy ranking · PASCAL-5^inovel-class mIoU (AVG): 2.3% improvement in average mIoUImproved
combine over-clustering, entropy ranking, dynamic reassignment, and self-training (ablation)Basic framework · PASCAL-5^inovel-class mIoU (AVG): outperforming the basic framework by 9.28%Improved
use over-clustering (ablation)Basic framework · COCO-20^inovel-class mIoU: Fold0 14.81%; Fold2 2.59%Mixed
add the EUMS framework (ablation)Basic framework · COCO-20^inovel-class mIoU (AVG): outperforms the baseline by 6.9%Improved
set the clean split ratio too low (0.33 or 0.50) (ablation)EUMS · PASCAL-5^inovel-class mIoUHurt
increase the clean split ratio beyond 0.67 (ablation)EUMS · PASCAL-5^inovel-class mIoUHurt
raise the online pseudo-label threshold to 0.9 (ablation)EUMS self-training · PASCAL-5^inovel-class mIoU: 3%Hurt
vary the online pseudo-label threshold within η < 0.5 (ablation)EUMS self-training · PASCAL-5^inovel-class mIoUNo clear effect
remove self-training by setting η = 1.0 (ablation)EUMS · PASCAL-5^inovel-class mIoUHurt

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