AI Paper Atlas

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

GeCo: Geometry-Consistent Regularization for Domain Generalized Semantic Segmentation

Qi Zang, Dong Zhao, Nan Pu, Wenjing Li, Zhun Zhong, Meng Wang

CVPR 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Domain-generalized segmentation

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
replace random Gaussian perturbation with curvature-guided perturbation (both with MSE) (ablation)GeCo · GTA5 -> Citys+BDD+Map (DINOv2-L)Avg mIoU: 62.8 -> 65.7Improved
replace random perturbation with curvature-guided perturbation (ablation)GeCo · GTA5 -> Citys+BDD+Map (EVA02-L)Avg mIoU: 64.0 -> 65.3Improved
add geodesic-based regularization on top of curvature-guided perturbation (replace MSE) (ablation)GeCo · GTA5 -> Citys+BDD+Map (DINOv2-L)Avg mIoU: 65.7 -> 67.7Improved
add geodesic-based regularization on top of curvature-guided perturbation (replace MSE) (ablation)GeCo · GTA5 -> Citys+BDD+Map (EVA02-L)Avg mIoU: 65.3 -> 66.3Improved
use geodesic regularization with random perturbation instead of MSE (ablation)random perturbation baseline · GTA5 -> Citys+BDD+Map (DINOv2-L)Avg mIoU: 62.8 -> 64.1Improved
add random perturbation with MSE consistency to full fine-tuning (ablation)EVA02-L · GTA5 -> Citys+BDD+MapAvg mIoU: 61.0 -> 64.0Improved
plug GeCo into ReinRein (DINOv2-L) · GTA5 -> Citys+BDD+MapAvg mIoU: 64.3 -> 66.3Improved
plug GeCo into FADAFADA (DINOv2-L) · GTA5 -> Citys+BDD+MapAvg mIoU: 66.1 -> 66.4Improved
GeCo vs SoMA in multi-source settingGeCo · GTA5+SYNTHIA -> Citys+BDD+MapAvg mIoU: 70.8 -> 71.1Mixed
add GeCo to RPL open-set pipeline instead of ReinRPL · Cityscapes -> MUADAP: 51.8 -> 58.6Improved

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