Research map / Detection and segmentation / CNN features
ArgoTweak: Towards Self-Updating HD Maps through Structured Priors
Lena Wild, Rafael Valencia, Patric Jensfelt
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 changed | Where | Measured | Effect |
|---|---|---|---|
| train with ArgoTweak priors instead of rule-based synthetic priors | LaneSegNet prior-aided · ArgoTweak real-world test (training prior comparison, no change modelling) | mAP: 76.7 -> 77.7 | Improved |
| add any map prior vs no prior | LaneSegNet · ArgoTweak real-world test (training prior comparison, no change modelling) | mAP: 39.4 -> 77.7 | Improved |
| continuous noise priors vs discrete modification priors | LaneSegNet prior-aided · ArgoTweak real-world test (training prior comparison, no change modelling) | mAP: 73.5 vs 71.5 | No clear effect |
| train on ArgoTweak instead of rule-based priors to reduce sim2real gap | Change-aware LaneSegNet · Sim2real gap mACC | Delta mACC: -36.0 -> -3.5 | Improved |
| add full atomic change annotations vs none (ablation) | Change-aware LaneSegNet · ArgoTweak test, annotation granularity | mAP: 77.7 -> 78.8 | Improved |
| add geometry change annotations to binary change labels (ablation) | Binary change head · ArgoTweak test, annotation granularity | mAPC: 46.9 -> 44.7 | Hurt |
| add geometry change annotations to binary change labels (ablation) | Binary change head · ArgoTweak test, annotation granularity | mACC: 66.5 -> 70.5 | Improved |
| add secondary change heads for marking and geometry (ablation) | Full atomic model · ArgoTweak test, annotation granularity | mAP: 78.3 -> 78.8 | Mixed |
| add secondary heads for marking and geometry (ablation) | Full atomic model · ArgoTweak test, annotation granularity | mACC: 64.3 -> 64.6 | No clear effect |
| cross-attention prior fusion vs prior tokens replacing queries | Prior encoding · ArgoTweak | qualitative | Improved |
Closest papers by meaning
- PriorMaskMap: Robust Online Vectorized Map Construction with Biased Priors (ECCV 2026 · similarity 0.87)
- Trust, but Verify: Cross-Modality Fusion for HD Map Change Detection (NeurIPS 2021 · similarity 0.87)
- SDTagNet: Leveraging Text-Annotated Navigation Maps for Online HD Map Construction (NeurIPS 2025 · similarity 0.85)
- The Change You Want To Detect: Semantic Change Detection In Earth Observation With Hybrid Data Generationf (CVPR 2025 · similarity 0.84)
- Localization Is All You Evaluate: Data Leakage in Online Mapping Datasets and How to Fix It (CVPR 2024 · similarity 0.84)
- Uncertainty-Instructed Structure Injection for Generalizable HD Map Construction (CVPR 2025 · similarity 0.83)
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.
