AI Paper Atlas

Research map / 3D vision / Depth and stereo

VectorReLoc: Reliable Vectorized SD Map Visual Re-localization with Contrastive Feature Alignment

Ziming Liu, Quanjie Xiang, Wang yun, wang yiting, chao wen, Zhuanjian XU, Leichen Wang, HAO SUN, Guangyu Gao

ECCV 2026 · 3D vision › Depth and stereo · approach: Scene coordinate regression for localization · 0 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 pretraining on the VectorReLoc dataset before fine-tuning (ablation)VectorReLoc · nuScenesposition recall @1m (%): 39.09 -> 65.79Improved
add pretraining on the VectorReLoc dataset before fine-tuning (ablation)VectorReLoc · Argoverse2position recall @1m (%): 49.86 -> 58.05Improved
add pretraining on the VectorReLoc dataset before fine-tuning (ablation)VectorReLoc · nuScenesEpos (m): 2.15 -> 1.36Improved
add contrastive joint training to regression-only (ablation)VectorReLoc · VectorReLoc dataset (real GNSS noise)position recall @1m (%): 22.96 -> 57.09Improved
add contrastive joint training to regression-only (ablation)VectorReLoc · VectorReLoc dataset (real GNSS noise)xy error (m): 3.0632 -> 1.3740Improved
add contrastive joint training to regression-only (ablation)VectorReLoc · VectorReLoc dataset (simulated GNSS noise)position recall @1m (%): 5.25 -> 19.27Improved
train contrastive and regression objectives jointly instead of pretrain-then-finetune (ablation)VectorReLoc · VectorReLoc dataset (real GNSS noise)position recall @1m (%): 52.40 -> 57.09Improved
add reliability filtering at threshold 0.9 (ablation)VectorReLoc · VectorReLoc datasetyaw error (deg): 0.3620 -> 0.2712Improved
train on synthetic Student-t offsets instead of real GNSS offsets (ablation)VectorReLoc · VectorReLoc dataset test splitEpos and 5 m recall: <=0.01 m in Epos and <=0.07 pp in 5 m recallNo clear effect
increase simulated training offset range from 1 m to 30 m (ablation)VectorReLoc · Argoverse2Epos (m): 1.04 -> 1.74 (1 m to 13 m); 1.67 at 30 mHurt
quantize to Int8 instead of Float32VectorReLoc · Horizon J6M deploymentposition recall @1m (%): 26.25 -> 17.35Hurt
quantize to Int16 instead of Float32VectorReLoc · Horizon J6M deploymentposition recall @1m (%): 26.25 -> 26.05No clear effect

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