AI Paper Atlas

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

VBLC: Visibility Boosting and Logit-Constraint Learning for Domain Adaptive Semantic Segmentation under Adverse Conditions

Mingjia Li, Binhui Xie, Shuang Li, Chi Harold Liu, Xinjing Cheng

AAAI 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Domain-generalized segmentation · 16 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 VBLC instead of FDADeepLab-v2 · ACDC testmIoU: 45.7 -> 47.8; +2.1%Improved
Use VBLC instead of DAFormerSegFormer · ACDC testmIoU: 55.3 -> 64.2; +8.9%Improved
Use VBLC instead of DAFormerSegFormer · FoggyCityscapes + RainCityscapesmIoU: 74.9 -> 79.8Improved
Add VBLC modules to SFADeformable DETR detection · FoggyCityscapes + RainCityscapesAP: 14.3 -> 15.6; 1.3 APImproved
Add conventional cross-entropy self-training on source-target blended images (ablation)SegFormer · Cityscapes → ACDCmIoU: 45.9 -> 52.6; +6.7%Improved
Add visibility boosting and boosted-image cross-entropy supervision (ablation)SegFormer with self-training · Cityscapes → ACDCmIoU: 52.6 -> 57.0; 4.4↑Improved
Use target logit-constraint learning instead of target cross-entropy losses (ablation)VBM + self-training · Cityscapes → ACDCmIoU: 57.0 -> 63.2; +6.2%Improved
Add source-domain logit-constraint learning (ablation)VBM + target logit-constraint learning · Cityscapes → ACDCmIoU: 63.2 -> 64.2; +1.0%Improved

Later papers that compared with this method

1 later accepted paper reports VBLC as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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