HeightMapNet

Explicit Height Modeling for End-to-End HD Map Learning

Introduction

This repository is an official implementation of HeightMapNet.

Getting Started

1. Environment

Step 1. Create conda environment and activate it.

conda create --name heightmapnet python=3.8 -y
conda activate heightmapnet

Step 2. Install PyTorch.

pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 torchaudio==0.9.0 -f https://download.pytorch.org/whl/torch_stable.html

Step 3. Install MMCV series.

# Install mmcv-series
pip install mmcv-full==1.6.0
pip install mmdet==2.28.2
pip install mmsegmentation==0.30.0
git clone https://github.com/open-mmlab/mmdetection3d.git
cd mmdetection3d
git checkout v1.0.0rc6 
pip install -e .

Step 4. Install other requirements.

pip install -r requirements.txt

2. Data Preparation

Step 1. Download NuScenes dataset to ./datasets/nuScenes.

Step 2. Generate annotation files for NuScenes dataset.

python tools/nuscenes_converter.py --data-root ./datasets/nuScenes --newsplit

3. Training and Validating

To train a model with 1 GPUs:

bash tools/dist_train.sh ${CONFIG} 1

To validate a model with 1 GPUs:

bash tools/dist_test_map.sh ${CONFIG} ${CEHCKPOINT} 1

Results

Results on NuScenes dataset

Method	Ap_divider	AP_boundary	AP_pedcrossing	mAP	Config	Download
HeightMapNet	62.8	60.4	54.3	59.1	config	ckpt

Citation

If you find our paper or codebase useful in your research, please cite our paper.

@misc{qiu2024heightmapnetexplicitheightmodeling,
      title={HeightMapNet: Explicit Height Modeling for End-to-End HD Map Learning}, 
      author={Wenzhao Qiu and Shanmin Pang and Hao zhang and Jianwu Fang and Jianru Xue},
      year={2024},
      eprint={2411.01408},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2411.01408}, 
}