Baike.dev
All toolsOnline toolsAI codingTrendingOpen sourceNewsSubmit
Log in
< Back to tools
P

P-MapNet

> 编程语言
Open source

Received by RAL

198 stars0 likes2 views
WebsiteGitHub

About

Received by RAL

P-MapNet: Far-seeing Map Generator Enhanced by both SDMap and HDMap Priors

[RAL] [Paper] [Project Page]

Abstract: Autonomous vehicles are gradually entering city roads today, with the help of high-definition maps (HDMaps). However, the reliance on HDMaps prevents autonomous vehicles from stepping into regions without this expensive digital infrastructure. This fact drives many researchers to study online HDMap construction algorithms, but the performance of these algorithms at far regions is still unsatisfying. We present P-MapNet, in which the letter P highlights the fact that we focus on incorporating map priors to improve model performance. Specifically, we exploit priors in both SDMap and HDMap. On one hand, we extract weakly aligned SDMap from OpenStreetMap, and encode it as an additional conditioning branch. Despite the misalignment challenge, our attention-based architecture adaptively attends to relevant SDMap skeletons and significantly improves performance. On the other hand, we exploit a masked autoencoder to capture the prior distribution of HDMap, which can serve as a refinement module to mitigate occlusions and artifacts. We benchmark on the nuScenes and Argoverse2 datasets. Through comprehensive experiments, we show that: (1) our SDMap prior can improve online map construction performance, using both rasterized (by up to +18.73 mIoU) and vectorized (by up to +8.50 mAP) output representations. (2) our HDMap prior can improve map perceptual metrics by up to 6.34%. (3) P-MapNet can be switched into different inference modes that covers different regions of the accuracy-efficiency trade-off landscape. (4) P-MapNet is a far-seeing solution that brings larger improvements on longer ranges.

Model

Results on nuScenes-val set

We provide results on nuScenes-val set.

Range Method M Div. Ped. Bound. mIoU Model Config
60 × 30 HDMapNet L+C 45.9 30.5 56.8 44.40 ckpt cfg
60 × 30 P-MapNet(SD+HD Prio.) L+C 54.2 41.3 63.7 53.07 ckpt cfg
120 × 60 HDMapNet L+C 53.6 37.8 57.1 49.50 ckpt cfg
120 × 60 P-MapNet(SD+HD Prio.) L+C 65.3 52.0 68.0 61.77 ckpt cfg
240 × 60 HDMapNet L+C 40.0 26.8 42.6 36.47 ckpt cfg
240 × 60 P-MapNet(SD+HD Prio.) L+C 53.0 42.6 54.2 49.93 ckpt cfg

The model weights under other settings can be downloaded at GoogleDrive or 百度云.

Getting Started

  • Installation
  • Train and Eval
  • visualization

TODO

  • Add Argoverse2 dataset model

Citation

If you found this paper or codebase useful, please cite our paper:

…

Issues· 0 open

View all issuesOpen on GitHub

No open issues yet, or sync has not completed.

> Tags

Python

No comments yet. Be the first to share.

> Details

PublishedAug 1, 2026
UpdatedSep 18, 2026
Category编程语言
PricingOpen source

> Related tools

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言