AI Paper Atlas

Research map / 3D vision / Depth and stereo

Topology Preserving Local Road Network Estimation From Single Onboard Camera Image

Yigit Baran Can, Alexander Liniger, Danda Pani Paudel, Luc Van Gool

CVPR 2022 · 3D vision › Depth and stereo · approach: Scene coordinate regression for localization · 39 citations (OpenAlex)

Check 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 minimal-cycle cover supervision to the transformer method (ablation)TR · NuScenesMC-F: 62.0 -> 62.5Improved
Add minimal-cycle cover supervision to the transformer method (ablation)TR · NuScenesI-Order: 0.800 -> 0.776Improved
Add minimal-cycle cover supervision to the transformer method (ablation)TR · ArgoverseMC-F: 57.4 -> 58.5Improved
Add minimal-cycle cover supervision to the transformer method (ablation)TR · ArgoverseI-Order: 0.893 -> 0.883Improved
Add minimal-cycle cover supervision to Polygon-RNN (ablation)PRNN · NuScenesMC-F: 45.9 -> 53.2Improved
Add minimal-cycle cover supervision to Polygon-RNN (ablation)PRNN · NuScenesI-Order: 0.894 -> 0.824Improved
Add minimal-cycle cover supervision to Polygon-RNN (ablation)PRNN · ArgoverseMC-F: 42.8 -> 49.5Improved
Add minimal-cycle cover supervision to Polygon-RNN (ablation)PRNN · ArgoverseI-Order: 1.09 -> 1.05Improved
Use ground-truth instead of estimated initial centerline points (ablation)Polygon-RNN · NuScenesMC-F: 53.2 -> 66.9Improved
Use ground-truth instead of estimated initial centerline points (ablation)Ours/PRNN · NuScenesMC-F: 53.2 -> 67.5Improved
Use ground-truth instead of estimated initial centerline points (ablation)Polygon-RNN · ArgoverseMC-F: 49.5 -> 61.0Improved
Use ground-truth instead of estimated initial centerline points (ablation)Ours/PRNN · ArgoverseMC-F: 49.5 -> 61.4Improved

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