Awesome

PPlanedet: A Toolkit for lane detection based on PaddlePaddle

In this project, we develop a toolkit for lane detection to facilitate research. Especially, PPlanedet is built upon Paddleseg which is a development Toolkit for segmentation based on PaddlePaddle.

If you do not have enough compute resource, we recommend that you can run our project at AiStudio, which can provide V100 with 32GB memory for free. We also open source the chinese version at AiStudio. Project link is here

News

Introduction

PPlanedet is developed for lane detection based on PaddlPaddle, which is a high performance Deep learning framework. The idea behind the pplanedet is to facilitate researchers who use PaddlePaddle to do research about lane detection. If you have any suggestions about our project, you can contact me.

Overview

Installation

step 1 Install PaddlePaddle(you can refer to official documentation)

conda create -n pplanedet python=3.8 -y
conda activate pplanedet
conda install paddlepaddle-gpu==2.3.2 cudatoolkit=10.2 --channel https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/Paddle/

step2 Install paddleseg

pip install paddleseg
# pip install git+"https://github.com/PaddlePaddle/PaddleSeg"

step3 Install requirements

cd PPlanedet
pip install -r requirements.txt

Data preparation

CULane

Download CULane. Then extract them to $CULANEROOT. Create link to data directory.

cd $LANEDET_ROOT
mkdir -p data
ln -s $CULANEROOT data/CULane

For CULane, you should have structure like this:

$CULANEROOT/driver_xx_xxframe    # data folders x6
$CULANEROOT/laneseg_label_w16    # lane segmentation labels
$CULANEROOT/list                 # data lists

Tusimple

Download Tusimple. Then extract them to $TUSIMPLEROOT. Create link to data directory.

cd $LANEDET_ROOT
mkdir -p data
ln -s $TUSIMPLEROOT data/tusimple

For Tusimple, you should have structure like this:

$TUSIMPLEROOT/clips # data folders
$TUSIMPLEROOT/lable_data_xxxx.json # label json file x4
$TUSIMPLEROOT/test_tasks_0627.json # test tasks json file
$TUSIMPLEROOT/test_label.json # test label json file

For Tusimple, the segmentation annotation is not provided, hence we need to generate segmentation from the json annotation.

python tools/generate_seg_tusimple.py --root $TUSIMPLEROOT
# this will generate seg_label directory

Getting Started

Training

For training, run

# training on single-GPU
export CUDA_VISIBLE_DEVICES=0
python train.py \
        --config configs/scnn/resnet50_tusimple.yaml \
        --do_eval \
        --use_vdl \
        --save_interval 1000 \
        --num_workers 4 \
        --save_dir output

# training on multi-GPU
export CUDA_VISIBLE_DEVICES=0,1,2,3
python -m paddle.distributed.launch train.py \
        --config configs/scnn/resnet50_tusimple.yaml \
        --do_eval \
        --use_vdl \
        --save_interval 1000 \
        --num_workers 4 \
        --save_dir output

Testing

For testing, run

python val.py \
       --config configs/scnn/resnet50_tusimple.yaml \
       --model_path /home/aistudio/181144/pplandet/output/best_model/model.pdparams #please change to your model path

Inference/Demo

python predict.py \
       --config configs/scnn/resnet50_tusimple.yaml \
       --model_path your model path
       --image_path your img path
       --custom_color 0 50 100 150 200

License

PPlanedet is released under the MIT license. We only allow you to use our project for academic uses.

Acknowledgement

• Thanks Paddleseg for codebase.
• Some codes are borrowed from lanedet.

Citation

If you find our project useful in your research, please consider citing:

@misc{PPlanedet,
    title={PPlanedet, A Toolkit for lane detection based on PaddlePaddle},
    author={Kunyang Zhou},
    howpublished = {\url{https://github.com/zkyseu/PPlanedet}},
    year={2022}
}

model reproduced in our project

@inproceedings{pan2018SCNN,  
  author = {Xingang Pan, Jianping Shi, Ping Luo, Xiaogang Wang, and Xiaoou Tang},  
  title = {Spatial As Deep: Spatial CNN for Traffic Scene Understanding},  
  booktitle = {AAAI Conference on Artificial Intelligence (AAAI)},  
  month = {February},  
  year = {2018}  
}