AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Informative Data Mining for One-Shot Cross-Domain Semantic Segmentation

Yuxi Wang, Jian Liang, Jun Xiao, Shuqi Mei, Yuran Yang, Zhaoxiang Zhang

ICCV 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adaptation for semantic segmentation · 14 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 sample-selected minimization (SSM) (ablation)DAFormer one-shot · GTA5->Cityscapes one-shotmIoU: 50.6 -> 52.0Improved
add prototype-based information maximization (PIM) to SSM (ablation)IDM · GTA5->Cityscapes one-shotmIoU: 52.0 -> 53.2Improved
add PatchMix to SSM (ablation)IDM · GTA5->Cityscapes one-shotmIoU: 52.0 -> 55.0Improved
use PatchMix instead of ClassMix (ablation)IDM · GTA5->Cityscapes one-shotmIoU: 55.9 -> 56.7Improved
add SSM when PatchMix and PIM present (ablation)IDM · GTA5->Cityscapes one-shotmIoU: 54.8 -> 56.7Improved
add style transfer plus prediction and similarity uncertainty selection (ablation)IDM sample selection · GTA5->Cityscapes one-shotiterations to reach mIoU: 20000 -> 300 iterations, 50.6 -> 52.0 mIoUImproved
naively reduce target data of UDA method to one shotDAFormer · GTA5->CityscapesmIoU: 68.3 -> 50.6Hurt
vary thresholds lambda_ent, lambda_sim, k (ablation)IDM · GTA5->CityscapesmIoU: 50.8 - 51.4 (lambda_ent)No clear effect
apply IDM to DeepLab-v2 backbone (ablation)DeepLab-v2 · GTA5->Cityscapes one-shotmIoU: 41.1 -> 45.2Improved

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