Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/Mrmoore98/VectorMapNet_code
This is the official code base of VectorMapNet (ICML 2023)
https://github.com/Mrmoore98/VectorMapNet_code
autonomous deep-learning map-learning
Last synced: 4 months ago
JSON representation
This is the official code base of VectorMapNet (ICML 2023)
- Host: GitHub
- URL: https://github.com/Mrmoore98/VectorMapNet_code
- Owner: Mrmoore98
- License: gpl-3.0
- Created: 2022-06-18T07:40:14.000Z (over 2 years ago)
- Default Branch: mian
- Last Pushed: 2023-10-07T13:02:49.000Z (over 1 year ago)
- Last Synced: 2024-08-01T03:42:30.726Z (7 months ago)
- Topics: autonomous, deep-learning, map-learning
- Language: Python
- Homepage: https://tsinghua-mars-lab.github.io/vectormapnet/
- Size: 151 KB
- Stars: 391
- Watchers: 29
- Forks: 54
- Open Issues: 32
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
- awesome-hd-map-construction - [code
README
# VectorMapNet_code
**VectorMapNet: End-to-end Vectorized HD Map Learning** ICML 2023
This is the official codebase of VectorMapNet
[Yicheng Liu](https://scholar.google.com/citations?user=vRmsgQUAAAAJ&hl=zh-CN), Yuantian Yuan, [Yue Wang](https://people.csail.mit.edu/yuewang/), [Yilun Wang](https://scholar.google.com.hk/citations?user=nUyTDosAAAAJ&hl=en/), [Hang Zhao](http://people.csail.mit.edu/hangzhao/)
**[[Paper](https://arxiv.org/pdf/2206.08920.pdf)] [[Project Page](https://tsinghua-mars-lab.github.io/vectormapnet/)]**
**Abstract:**
Autonomous driving systems require High-Definition (HD) semantic maps to navigate around urban roads. Existing solutions approach the semantic mapping problem by offline manual annotation, which suffers from serious scalability issues. Recent learning-based methods produce dense rasterized segmentation predictions to construct maps. However, these predictions do not include instance information of individual map elements and require heuristic post-processing to obtain vectorized maps. To tackle these challenges, we introduce an end-to-end vectorized HD map learning pipeline, termed VectorMapNet. VectorMapNet takes onboard sensor observations and predicts a sparse set of polylines in the bird's-eye view. This pipeline can explicitly model the spatial relation between map elements and generate vectorized maps that are friendly to downstream autonomous driving tasks. Extensive experiments show that VectorMapNet achieve strong map learning performance on both nuScenes and Argoverse2 dataset, surpassing previous state-of-the-art methods by 14.2 mAP and 14.6mAP. Qualitatively, VectorMapNet is capable of generating comprehensive maps and capturing fine-grained details of road geometry. To the best of our knowledge, VectorMapNet is the first work designed towards end-to-end vectorized map learning from onboard observations.
**Questions/Requests:**
Please file an [issue](https://github.com/Tsinghua-MARS-Lab/vecmapnet/issues) or send an email to [Yicheng]([email protected]).
## Bibtex
If you found this paper or codebase useful, please cite our paper:
```
@inproceedings{liu2022vectormapnet,
title={VectorMapNet: End-to-end Vectorized HD Map Learning},
author={Liu, Yicheng and Yuantian, Yuan and Wang, Yue and Wang, Yilun and Zhao, Hang},
booktitle={International conference on machine learning},
year={2023},
organization={PMLR}
}
```
# Run VectorMapNet
## Note
## 0. Environment
Set up environment by following this [script](env.md)
## 1. Prepare your dataset
Store your data with following structure:
```
root
|--datasets
|--nuScenes
|--Argoverse2(optional)
```
### 1.1 Generate annotation files
#### Preprocess nuScenes
```
python tools/data_converter/nuscenes_converter.py --data-root your/dataset/nuScenes/
```
## 2. Evaluate VectorMapNet
### Download Checkpoint
| Method | Modality | Config | Checkpoint |
|--------------|-------------|--------|------------|
| VectorMapNet | Camera only | [config](configs/vectormapnet.py) | [model link](https://drive.google.com/file/d/1ccrlZ2HrFfpBB27kC9DkwCYWlTUpgmin/view?usp=sharing) |
### Train VectorMapNet
In single GPU
```
python tools/train.py configs/vectormapnet.py
```
For multi GPUs
```
bash tools/dist_train.sh configs/vectormapnet.py $num_gpu
```
### Do Evaluation
In single GPU
```
python tools/test.py configs/vectormapnet.py /path/to/ckpt --eval name
```
For multi GPUs
```
bash tools/dist_test.sh configs/vectormapnet.py /path/to/ckpt $num_gpu --eval name
```
### Expected Results
| $AP_{ped}$ | $AP_{divider}$ | $AP_{boundary}$ | mAP |
|--------------|----------------|-----------------|-------|
| 39.8 | 47.7 | 38.8 | 42.1 |