AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Online Map Vectorization for Autonomous Driving: A Rasterization Perspective

Gongjie Zhang, Jiahao Lin, Shuang Wu, Yilin Song, Zhipeng Luo, Yang Xue, Shijian Lu, Zuoguan Wang

NeurIPS 2023 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 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
Add MapVR rasterized supervision to MapTRMapTR · nuScenes Map (basic) val, camera, 24 epochsAP_raster avg.: 24.3 -> 31.2Improved
Add MapVR rasterized supervision to MapTRMapTR · nuScenes Map (basic) val, camera, 110 epochsAP_raster avg.: 35.0 -> 38.5Improved
Add MapVR rasterized supervision to MapTRMapTR · nuScenes Map (extended) valAP_raster avg.: 21.1 -> 27.8Improved
Add MapVR rasterized supervision to MapTRMapTR · Argoverse2 valAP_raster avg.: 26.2 -> 30.1Improved
Add MapVR rasterized supervision to MapTRMapTR · 6V-mini-v0.4AP_raster lane: 41.3 -> 50.8Improved
Use coarse 64x32 rasterization instead of the default 256x128 (ablation)MapTR + MapVR · nuScenes Map (basic) valmAP_raster: 31.2 -> 21.5Hurt
Set line rasterization softness τ to 0.5 rather than the default 2.0 (ablation)MapTR + MapVR · nuScenes Map (basic) val, dividermAP_raster: 33.1 -> 29.2Hurt
Add self-based direction regularization on predicted points (ablation)MapTR + MapVR · nuScenes Map (basic) valmAP_raster: 29.5 -> 31.2Improved
Use ground-truth point directions for regularization instead of self-based regularization (ablation)MapTR + MapVR · nuScenes Map (basic) valmAP_raster: 31.2 -> 29.3Hurt
Use geometry-aware line and polygon rasterization rather than rendering every element as a line (ablation)MapTR + MapVR · nuScenes Map (basic) val, pedestrian crossingmAP_raster: 21.8 -> 37.5Improved
Apply rasterized supervision directly to vector predictions rather than using a parallel segmentation branch (ablation)MapTR · nuScenes Map (basic) valmAP_raster: 26.7 -> 31.2Improved

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